DETAILED ACTION
This Office Action is in response to communications filed on March 27th, 2026 for Application No. 18/103,606, in which claims 1-20 are presented for examination. The amendments filed on March 27th, 2026 have been entered, where claims 1-2, 6, 12-13, and 18-20 are amended.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly
point out and distinctly claim the subject matter which the inventor or a joint inventor regards
as the invention.
Regarding Claim 1, the claim recites the terms “educational journey” (ln. 2, 23, 25, 26, and 28), “surfaced and not washed away” (ln. 12-13), “hindering” (ln. 24), and “promoting” (ln. 25), which are relative terms that render the claim indefinite. These terms are not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree of “journey”, “surfaced and not washed away”, “hindering”, or “promoting” that is required, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. As a result, it is not clear what qualifies as “points in the educational journey” (ln. 23) or “progress in the educational journey” (ln. 24-26 and 28) because it is unclear what qualifies as an “educational journey”. Additionally, it is not clear what degree of “maintaining separation” (ln. 11) is needed “such that points unique to a smaller-volume unstructured data source are surfaced and not washed away by a larger-volume unstructured data source” (ln. 12-13). Furthermore, it is not clear what qualifies as “points” (ln. 23) because it is unclear what qualifies as “points that are associated with hindering the particular student's progress in the educational journey and (ii) points that are associated with promoting the particular student's progress in the educational journey” (ln. 23 -26). Therefore, the claim is rejected. The claim should be amended to clarify the scopes of “educational journey”, “surfaced and not washed away”, “hindering”, and “promoting”.
Regarding Claim 2-3, the claims are rejected because they are dependent on a rejected claim.
Regarding Claim 4, the claim recites “journey” (ln. 5), which is indefinite for substantially the same reasons as those articulated in regard to the rejection of Claim 1. As a result, the claim is similarly rejected and should be amended in a similar manner. Additionally, the claim is rejected because it is dependent on a rejected claim.
Regarding Claims 5-11, the claims are rejected because they are dependent on a rejected claim.
Regarding Claim 12, the claim recites “educational journey” (ln. 28, 30, 31, and 38), “surfaced and not washed away” (ln. 19), “hindering” (ln. 29), and “promoting” (ln. 30), which are indefinite for substantially the same reasons as those articulated in regard to the rejection of Claim 1. As a result, the claim is similarly rejected and should be amended in a similar manner.
Additionally, the claim recites “the educational journey” (ln. 28, 29-30, 31, and 38). However, there is insufficient antecedent basis for this limitation in the claim. Therefore, the claim is rejected. The claim should be amended to remedy this issue, such as by replacing the initial recitation of “the educational journey” with “an educational journey”.
Regarding Claims 13 - 18, the claims are rejected because they are dependent on a rejected claim.
Regarding Claim 19, the claim recites the term “educational journey” (ln. 27, 29, 31, and 34), “surfaced and not washed away” (ln. 18), “hindering” (ln. 28), and “promoting” (ln. 29), which are indefinite for substantially the same reasons as those articulated in regard to the rejection of Claim 1. As a result, the claim is similarly rejected and should be amended in a similar manner.
Additionally, the claim recites “the educational journey” (ln. 27, 29, 31, and 34), which is indefinite for substantially the same reasons as those articulated in regard to the rejection of Claim 12. As a result, the claim is similarly rejected and should be amended in a similar manner.
Regarding Claim 20, the claim recites “the friction points” (ln. 2-3 and 4) and “the achievement points” (ln. 3-4). There is insufficient antecedent basis for this limitation in the claim. Specifically, Claim 19, which Claim 20 depends upon, was amended to remove recitations of the friction points and achievement points (see amended claim 19). As a result, it is not clear what recitations of “the friction points” and “the achievement points” in Claim 20 are referring to. Additionally, the terms “friction” (ln. 2-3 and 4) and “achievement” (ln. 3-4) are relative terms that render the claim indefinite. These terms are not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree of “friction” or “achievement” that is required, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. As a result, it is not clear what qualifies as “points”. Therefore, the claim is indefinite. As a result, the claim is rejected. The claim should be amended to correct this issue, such as by removing recitations of the friction points and the achievement points in a manner similar to Claim 19.
Additionally, the claim is rejected because it is dependent on a rejected claim.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract ideas without significantly more.
Regarding Claim 1:
Step 1: Claim 1 is a process claim. Therefore, Claims 1-11 are directed to a statutory category of eligible subject matter.
Step 2A Prong 1: If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the "Mental Processes" grouping of abstract ideas. Here, steps of the claimed subject matter are mental processes. Specifically, the claim recites
“A method of generating a predictive mapping of an educational journey of a student at an education institution, the method comprising . . . ” (mental process – amounts to exercising judgment to form an opinion on a predictive mapping, based on observed or known student information, which may be aided by pen and paper);
“normalizing the unstructured data to classify the unstructured data” (mental process – amounts to exercising judgment to normalize known or observed data for classification, which may be aided by pen and paper);
“maintaining separation of the plurality of unstructured data sources . . . such that points unique to a smaller-volume unstructured data source are surfaced and not washed away by a larger-volume unstructured data source, including (i) maintaining the source-separated” (mental process – amounts to exercising judgment to form an opinion that separate information should be maintained as distinct from one another, with reference to a particular goal of surfacing information to prevent it from being washed out, which may be aided by pen and paper)
“to classify the unstructured data into a plurality of classifications . . . classifying . . . the unstructured data into the plurality of classifications” (mental process – amounts to exercising judgment to form an opinion on classifications, based on observed or known data, which may be aided by pen and paper);
“based on the classifications of the unstructured data, identifying at least one of a plurality of points in the educational journey, the points including (i) points that are associated with hindering the particular student's progress in the educational journey and (ii) points that are associated with promoting the particular student's progress in the educational journey” (mental process – amounts to exercising judgment to evaluate information to form an opinion on parts of information that belong to a category, which may be aided by pen and paper);
“using the points, generating prediction information of the particular student's progress in the educational journey, the prediction information” (mental process – amounts to exercising judgment to evaluate information to form an opinion on a prediction, which may be aided by pen and paper);
“generating a journey map that includes the prediction information and includes one or more of the points selected based on a magnitude of correlation between (i) the points and (ii) a predicted event indicated by the prediction information, wherein the selected points include points having a magnitude of correlation exceeding a predetermined threshold or include a top predetermined number of points by the magnitude of correlation” (mental process – amounts to exercising judgement to form an opinion on a journey map, with reference to known or observed information, which are selected for based on a preference for information associated with a top number of values or above a known threshold, which may be aided by pen and paper); and
“wherein generating the journey map comprises including, in the journey map, extracted portions of the unstructured data that are extracted by identifying phrases used to categorize the unstructured data into the points” (mental process – amounts to exercising judgement to identify portions of known or observed data, with reference to aspects of the data used for categorization, which may be aided by pen and paper).
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“receiving, over a computer network, from a database of student records, unstructured data about a particular student of the educational institution . . . and transmitting the prediction information and the journey map over the computer network to an administrator machine” (receiving and transmitting data amounts to insignificant extra-solution activity because the receiving and transmission of data is incidental to the claimed subject matter);
“consistent with a machine learning classification model . . . for input into the machine learning classification model . . . wherein training the machine learning classification model comprises training the machine learning classification model using . . . training data sets . . . during the training . . . and (ii) training using the . . . data sets . . . training data sets. . . using the machine learning classification model . . . being consistent with a machine learning prediction model” (mere instructions to apply the exception using generic computer components does not provide an inventive concept); and
“source-separated . . . respectively corresponding to a plurality of unstructured data sources, including . . . training data sets as separate training data sets corresponding to the plurality of unstructured data sources . . . source-separated . . . without combining the source-separated training data sets into a single training data set” (amounts to merely generally linking the use of the judicial exception to a particular technological environment or field of use, which do not impose any meaningful limits on practicing the abstract idea).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“receiving, over a computer network, from a database of student records, unstructured data about a particular student of the educational institution . . . and transmitting the prediction information and the journey map over the computer network to an administrator machine” (transmitting data over a network is well‐understood, routine, and conventional, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362; OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014); therefore the limitation, which is recited with a high level of generality, remains insignificant extra-solution activity even upon reconsideration);
“consistent with a machine learning classification model . . . for input into the machine learning classification model . . . wherein training the machine learning classification model comprises training the machine learning classification model using . . . training data sets . . . during the training . . . and (ii) training using the . . . data sets . . . training data sets. . . using the machine learning classification model . . . being consistent with a machine learning prediction model” (mere instructions to apply the exception using generic computer components does not provide an inventive concept); and
“source-separated . . . respectively corresponding to a plurality of unstructured data sources, including . . . training data sets as separate training data sets corresponding to the plurality of unstructured data sources . . . source-separated . . . without combining the source-separated training data sets into a single training data set” (merely generally linking the use of the judicial exception to a particular technological environment or field of use does not provide an inventive concept).
For the reasons above, Claim 1 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claims 2-11. The additional limitations of the dependent claims are addressed below.
Regarding Claim 2:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 2 depends on. Here, the claim recites additional elements that are mental processes. Specifically, the claim recites
“wherein the prediction information is included in a journey map comprising an indication of a plurality of the points, including summary information for the plurality of the points” (mental process – amounts to exercising judgment to form an opinion, where the opinion includes specific indication and summary information, which may be aided by pen and paper).
Step 2A Prong 2 & Step 2B: There are no elements left for consideration of implementation within a practical application or for consideration of significantly more.
Accordingly, Claim 2 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 3:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 3 depends on.
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“wherein the unstructured data comprises at least one of mentor notes, email interactions, assessment responses, instructor notes, social media posts, course surveys, personality test responses, or aptitude test responses” (amounts to merely generally linking the use of the judicial exception to a particular technological environment or field of use, which do not impose any meaningful limits on practicing the abstract idea).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“wherein the unstructured data comprises at least one of mentor notes, email interactions, assessment responses, instructor notes, social media posts, course surveys, personality test responses, or aptitude test responses” (merely generally linking the use of the judicial exception to a particular technological environment or field of use does not provide an inventive concept).
Accordingly, Claim 3 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 4:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 4 depends on. Here, the claim recites additional elements that are mental processes. Specifically,
“generating the prediction information of the particular student's progress in the educational journey” (mental process – amounts to forming an opinion based on known or observed information, which may be aided by pen and paper).
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“receiving over the computer network, structured data” (providing data amounts to insignificant extra-solution activity because the transmission of data is incidental to the claimed subject matter) and
“wherein the structured data is used in the machine learning prediction model in” (mere instructions to apply the exception using generic computer components does not provide an inventive concept).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“receiving over the computer network, structured data” (transmitting data is well‐understood, routine, and conventional, see generally Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362; see also buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014); therefore the limitation, which is recited with a high level of generality, remains insignificant extra-solution activity even upon reconsideration) and
“wherein the structured data is used in the machine learning prediction model in” (amounts to mere instructions to apply the judicial exception on generic and unspecialized computer components, which do not impose any meaningful limits on practicing the abstract idea).
Accordingly, Claim 4 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 5:
Step 2A Prong 1: See the rejection of Claim 4 above, which Claim 5 depends on.
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“wherein the structured data and unstructured data comprises at least one of mentor notes, email interactions, helpdesk tickets, program information, assigned grades, discipline write ups, assessment responses, financial aid status, transferred credits, academic resource interactions, instructor notes, marketing data, social media posts, governmental body reports, course survey responses, personality test responses, or aptitude test responses” (amounts to merely generally linking the use of the judicial exception to a particular technological environment or field of use, which do not impose any meaningful limits on practicing the abstract idea).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“wherein the structured data and unstructured data comprises at least one of mentor notes, email interactions, helpdesk tickets, program information, assigned grades, discipline write ups, assessment responses, financial aid status, transferred credits, academic resource interactions, instructor notes, marketing data, social media posts, governmental body reports, course survey responses, personality test responses, or aptitude test responses” (merely generally linking the use of the judicial exception to a particular technological environment or field of use does not provide an inventive concept).
Accordingly, Claim 5 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 6:
Step 2A Prong 1: See the rejection of Claim 4 above, which Claim 6 depends on. Here, the claim recites additional elements that are mental processes. Specifically, the claim recites
“wherein the structured data is used to generate points” (mental process – amounts to exercising judgment to form an opinion, based on known or observed information, which may be aided by pen and paper).
Step 2A Prong 2 & Step 2B: There are no elements left for consideration of implementation within a practical application or for consideration of significantly more.
Accordingly, Claim 6 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 7:
Step 2A Prong 1: See the rejection of Claim 4 above, which Claim 7 depends on.
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“wherein the structured data is used to refine the machine learning prediction model” (mere instructions to apply the exception using generic computer components does not provide an inventive concept).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“wherein the structured data is used to refine the machine learning prediction model” (amounts to mere instructions to apply the judicial exception on generic and unspecialized computer components, which do not impose any meaningful limits on practicing the abstract idea).
Accordingly, Claim 7 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 8:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 8 depends on.
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“wherein the unstructured data, based on the classifications, is used to refine the machine learning prediction model” (mere instructions to apply the exception using generic computer components does not provide an inventive concept).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“wherein the unstructured data, based on the classifications, is used to refine the machine learning prediction model” (amounts to mere instructions to apply the judicial exception on generic and unspecialized computer components, which do not impose any meaningful limits on practicing the abstract idea).
Accordingly, Claim 8 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 9:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 9 depends on.
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“wherein the machine learning classification model comprises a natural language processing model” (mere instructions to apply the exception using generic computer components does not provide an inventive concept).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“wherein the machine learning classification model comprises a natural language processing model” (amounts to mere instructions to apply the judicial exception on generic and unspecialized computer components, which do not impose any meaningful limits on practicing the abstract idea).
Accordingly, Claim 9 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 10:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 10 depends on.
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“wherein the machine learning prediction model comprises a root cause analysis model” (mere instructions to apply the exception using generic computer components does not provide an inventive concept).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“wherein the machine learning prediction model comprises a root cause analysis model” (amounts to mere instructions to apply the judicial exception on generic and unspecialized computer components, which do not impose any meaningful limits on practicing the abstract idea).
Accordingly, Claim 10 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 11:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 11 depends on. Here, the claim recites additional elements that are mental processes. Specifically,
“generate intervention output suggesting action to alter the generated prediction information” (mental process – amounts to forming an opinion based on known or observed information, which may be aided by pen and paper).
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“transmitting the intervention output over the computer network to the administrator machine” (providing data amounts to insignificant extra-solution activity because the transmission of data is incidental to the claimed subject matter) and
“using the machine learning prediction model to” (mere instructions to apply the exception using generic computer components does not provide an inventive concept).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“transmitting the intervention output over the computer network to the administrator machine” (transmitting data is well‐understood, routine, and conventional, see generally Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362; see also buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014); therefore the limitation, which is recited with a high level of generality, remains insignificant extra-solution activity even upon reconsideration) and
“using the machine learning prediction model to” (amounts to mere instructions to apply the judicial exception on generic and unspecialized computer components, which do not impose any meaningful limits on practicing the abstract idea).
Accordingly, Claim 11 is rejected as being directed to an abstract idea without significantly more.
Regarding Claim 12:
Step 1: Claim 12 is a machine claim. Therefore, Claims 12-18 are directed to a statutory category of eligible subject matter.
Step 2A Prong 1: If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the "Mental Processes" grouping of abstract ideas. Here, steps of the claimed subject matter are mental processes. Specifically, the claim recites
“normalize the unstructured data to classify the unstructured data” (mental process – amounts to exercising judgment to normalize known or observed data for classification, which may be aided by pen and paper);
“maintaining separation of the plurality of unstructured data sources . . . such that points unique to a smaller-volume unstructured data source are surfaced and not washed away by a larger-volume unstructured data source, including (i) maintaining the source-separated” (mental process – amounts to exercising judgment to form an opinion that separate information should be maintained as distinct from one another, with reference to a particular goal of surfacing information to prevent it from being washed out, which may be aided by pen and paper)
“to classify the unstructured data into a plurality of classifications . . . classify . . . the unstructured data into the plurality of classifications” (mental process – amounts to exercising judgment to form an opinion on classifications, based on observed or known data, which may be aided by pen and paper);
“based on the classifications of the unstructured data, identify at least one of a plurality of points in the educational journey, the points including (i) points that are associated with hindering the particular student's progress in the educational journey and (ii) points that are associated with promoting the particular student's progress in the educational journey” (mental process – amounts to exercising judgment to evaluate information to form an opinion on parts of information that belong to a category, which may be aided by pen and paper);
“using the points, generate prediction information of the particular student's progress in the educational journey, the prediction information” (mental process – amounts to exercising judgment to evaluate information to form an opinion on a prediction, which may be aided by pen and paper);
“generate a journey map that includes the prediction information and includes one or more of the points selected based on a magnitude of correlation between (i) the points and (ii) a predicted event indicated by the prediction information, wherein the selected points include points having a magnitude of correlation exceeding a predetermined threshold or include a top predetermined number of points by the magnitude of correlation” (mental process – amounts to exercising judgement to form an opinion on a journey map, with reference to known or observed information, which are selected for based on a preference for information associated with a top number of values or above a known threshold, which may be aided by pen and paper); and
“wherein generating the journey map comprises including, in the journey map, extracted portions of the unstructured data that are extracted by identifying phrases used to categorize the unstructured data into the points” (mental process – amounts to exercising judgement to identify portions of known or observed data, with reference to aspects of the data used for categorization, which may be aided by pen and paper).
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“receive, over a computer network, from a database of student records, unstructured data about a particular student of the educational institution . . . and transmit the prediction information and the journey map over the computer network to an administrator machine” (receiving and transmitting data amounts to insignificant extra-solution activity because the receiving and transmission of data is incidental to the claimed subject matter);
“A computing system comprising: one or more processors; and one or more computer-readable media having stored thereon instructions that are executable by the one or more processors; network hardware configured to . . . a categorization engine comprising a machine learning classification model, implemented by the one or more processors and the instructions, configured to: . . . consistent with a machine learning classification model . . . for input into the machine learning classification model . . . wherein training the machine learning classification model comprises training the machine learning classification model using . . . training data sets . . . during the training . . . and (ii) training using the . . . data sets . . . training data sets. . . using the machine learning classification model . . . a prediction engine comprising a machine learning prediction model, implemented by the one or more processors and the instructions, configured to: . . . being consistent with a machine learning prediction model . . . wherein the network hardware is configured to” (mere instructions to apply the exception using generic computer components does not provide an inventive concept); and
“source-separated . . . respectively corresponding to a plurality of unstructured data sources, including . . . training data sets as separate training data sets corresponding to the plurality of unstructured data sources . . . source-separated . . . without combining the source-separated training data sets into a single training data set” (amounts to merely generally linking the use of the judicial exception to a particular technological environment or field of use, which do not impose any meaningful limits on practicing the abstract idea).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“receive, over a computer network, from a database of student records, unstructured data about a particular student of the educational institution . . . and transmit the prediction information and the journey map over the computer network to an administrator machine” (transmitting data over a network is well‐understood, routine, and conventional, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362; OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014); therefore the limitation, which is recited with a high level of generality, remains insignificant extra-solution activity even upon reconsideration);
“A computing system comprising: one or more processors; and one or more computer-readable media having stored thereon instructions that are executable by the one or more processors; network hardware configured to . . . a categorization engine comprising a machine learning classification model, implemented by the one or more processors and the instructions, configured to: . . . consistent with a machine learning classification model . . . for input into the machine learning classification model . . . wherein training the machine learning classification model comprises training the machine learning classification model using . . . training data sets . . . during the training . . . and (ii) training using the . . . data sets . . . training data sets. . . using the machine learning classification model . . . a prediction engine comprising a machine learning prediction model, implemented by the one or more processors and the instructions, configured to: . . . being consistent with a machine learning prediction model . . . wherein the network hardware is configured to” (mere instructions to apply the exception using generic computer components does not provide an inventive concept); and
“source-separated . . . respectively corresponding to a plurality of unstructured data sources, including . . . training data sets as separate training data sets corresponding to the plurality of unstructured data sources . . . source-separated . . . without combining the source-separated training data sets into a single training data set” (merely generally linking the use of the judicial exception to a particular technological environment or field of use does not provide an inventive concept).
For the reasons above, Claim 12 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claims 13-18. The additional limitations of the dependent claims are addressed below.
Regarding Claim 13, the claim recites limitations that are all substantially the same as limitations of Claim 2, in the form of a machine. The claim is also directed to performing mental processes without integration into a practical component or significantly more.
Accordingly, Claim 13 is rejected under the same rationale.
Regarding Claim 14, the claim recites limitations that are all substantially the same as limitations of Claim 3, in the form of a machine. The claim is also directed to performing mental processes without integration into a practical component or significantly more.
Accordingly, Claim 14 is rejected under the same rationale.
Regarding Claim 15, the claim recites limitations that are all substantially the same as limitations of Claim 8, in the form of a machine. The claim is also directed to performing mental processes without integration into a practical component or significantly more.
Accordingly, Claim 15 is rejected under the same rationale.
Regarding Claim 16, the claim recites limitations that are all substantially the same as limitations of Claim 9, in the form of a machine. The claim is also directed to performing mental processes without integration into a practical component or significantly more.
Accordingly, Claim 16 is rejected under the same rationale.
Regarding Claim 17, the claim recites limitations that are all substantially the same as limitations of Claim 10, in the form of a machine. The claim is also directed to performing mental processes without integration into a practical component or significantly more.
Accordingly, Claim 17 is rejected under the same rationale.
Regarding Claim 18, the claim recites limitations that are all substantially the same as limitations of Claim 11, in the form of a machine. The claim is also directed to performing mental processes without integration into a practical component or significantly more.
Accordingly, Claim 18 is rejected under the same rationale.
Regarding Claim 19:
Step 1: Claim 19 is a machine claim. Therefore, Claims 19-20 are directed to a statutory category of eligible subject matter.
Step 2A Prong 1: If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the "Mental Processes" grouping of abstract ideas. Here, steps of the claimed subject matter are mental processes. Specifically, the claim recites
“normalizing the unstructured data to classify the unstructured data” (mental process – amounts to exercising judgment to normalize known or observed data for classification, which may be aided by pen and paper);
“maintaining separation of the plurality of unstructured data sources . . . such that points unique to a smaller-volume unstructured data source are surfaced and not washed away by a larger-volume unstructured data source, including (i) maintaining the source-separated” (mental process – amounts to exercising judgment to form an opinion that separate information should be maintained as distinct from one another, with reference to a particular goal of surfacing information to prevent it from being washed out, which may be aided by pen and paper)
“to classify the unstructured data into a plurality of classifications . . . classifying . . . the unstructured data into the plurality of classifications” (mental process – amounts to exercising judgment to form an opinion on classifications, based on observed or known data, which may be aided by pen and paper);
“based on the classifications of the unstructured data, identify at least one of a plurality of points in the educational journey, the points including (i) points that are associated with hindering the particular student's progress in the educational journey and (ii) points that are associated with promoting the particular student's progress in the educational journey” (mental process – amounts to exercising judgment to evaluate information to form an opinion on parts of information that belong to a category, which may be aided by pen and paper);
“using the points, generate prediction information of the particular student's progress in the educational journey, the prediction information” (mental process – amounts to exercising judgment to evaluate information to form an opinion on a prediction, which may be aided by pen and paper);
“predictive journey mapping . . . generating a journey map that includes the prediction information and includes one or more of the points selected based on a magnitude of correlation between (i) the points and (ii) a predicted event indicated by the prediction information, wherein the selected points include points having a magnitude of correlation exceeding a predetermined threshold or include a top predetermined number of points by the magnitude of correlation” (mental process – amounts to exercising judgement to form an opinion on a journey map, with reference to known or observed information, which are selected for based on a preference for information associated with a top number of values or above a known threshold, which may be aided by pen and paper); and
“wherein generating the journey map comprises including, in the journey map, extracted portions of the unstructured data that are extracted by identifying phrases used to categorize the unstructured data into the points” (mental process – amounts to exercising judgement to identify portions of known or observed data, with reference to aspects of the data used for categorization, which may be aided by pen and paper).
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
The claim recites the additional elements:
“receive, over a computer network, from a database of student records, unstructured data about a particular student of the educational institution . . . and transmit the prediction information and over the computer network to an administrator machine” (receiving and transmitting data amounts to insignificant extra-solution activity because the receiving and transmission of data is incidental to the claimed subject matter);
“A computing system comprising one or more processors; and one or more computer-readable media having stored thereon instructions that are executable by the one or more processors to configure the computer system to perform . . . including instructions that are executable to configure the computer system to perform at least the following: . . . consistent with a machine learning classification model . . . for input into the machine learning classification model . . . wherein training the machine learning classification model comprises training the machine learning classification model using . . . training data sets . . . during the training . . . and (ii) training using the . . . data sets . . . training data sets. . . using the machine learning classification model . . . being consistent with a machine learning prediction model” (mere instructions to apply the exception using generic computer components does not provide an inventive concept); and
“source-separated . . . respectively corresponding to a plurality of unstructured data sources, including . . . training data sets as separate training data sets corresponding to the plurality of unstructured data sources . . . source-separated . . . without combining the source-separated training data sets into a single training data set” (amounts to merely generally linking the use of the judicial exception to a particular technological environment or field of use, which do not impose any meaningful limits on practicing the abstract idea).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The claim recites the additional elements:
“receive, over a computer network, from a database of student records, unstructured data about a particular student of the educational institution . . . and transmit the prediction information and over the computer network to an administrator machine” (transmitting data over a network is well‐understood, routine, and conventional, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362; OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014); therefore the limitation, which is recited with a high level of generality, remains insignificant extra-solution activity even upon reconsideration);
“A computing system comprising one or more processors; and one or more computer-readable media having stored thereon instructions that are executable by the one or more processors to configure the computer system to perform . . . including instructions that are executable to configure the computer system to perform at least the following: . . . consistent with a machine learning classification model . . . for input into the machine learning classification model . . . wherein training the machine learning classification model comprises training the machine learning classification model using . . . training data sets . . . during the training . . . and (ii) training using the . . . data sets . . . training data sets. . . using the machine learning classification model . . . being consistent with a machine learning prediction model” (mere instructions to apply the exception using generic computer components does not provide an inventive concept); and
“source-separated . . . respectively corresponding to a plurality of unstructured data sources, including . . . training data sets as separate training data sets corresponding to the plurality of unstructured data sources . . . source-separated . . . without combining the source-separated training data sets into a single training data set” (merely generally linking the use of the judicial exception to a particular technological environment or field of use does not provide an inventive concept).
For the reasons above, Claim 19 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claim 20. The additional limitations of the dependent claim are addressed below.
Regarding Claim 20:
Step 2A Prong 1: See the rejection of Claim 19 above, which Claim 20 depends on. Here, the claim recites additional elements that are mental processes. Specifically, the claim recites
“wherein the prediction information is included in journey map comprising an indication of a plurality of the friction points or a plurality of the achievement points, including summary information for the plurality of the friction points or the plurality of the achievement points” (mental process – amounts to exercising judgment to form an opinion, where the opinion includes specific indication and summary information, which may be aided by pen and paper).
Step 2A Prong 2 & Step 2B: There are no elements left for consideration of implementation within a practical application or for consideration of significantly more.
Accordingly, Claim 20 is rejected as being directed to an abstract idea without significantly more.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Kil et al. (hereinafter Kil) (Patent Pub. No. US 2017/0256172 A1) in view of Kumar et al. (hereinafter Kumar) (Patent Pub. No. US 2020/0012930 A1) and Sajjad et al. (hereinafter Sajjad) (“Neural Machine Translation Training in a Multi-Domain Scenario”).
Regarding Claim 1, Kil teaches a method of generating a predictive mapping of an educational journey of a student at an education institution, the method comprising (Para. [0005], “A student data-to-insight-to-action-to-learning analytics method in accordance with an embodiment of the invention comprises computing student success predictions, student engagement predictions, and student impact predictions to interventions ”, where the “predictions” are incorporated into a predictive mapping, see Fig. 14, of an educational journey of a student at an educational institution; see also Para. [0039], “By virtue of having a different subset of top predictors during various stages of a student's academic journey, higher-education (HE) institutions can develop more timely and context-aware student outreach programs and policies”):
receiving, over a computer network, from a database of student records, unstructured data about a particular student of the educational institution (Para. [0166], “one implementation of the micro intervention delivery subsystem 106 . . . uses incoming event data streams from multiple student event data sources, such as, but not limited to, SIS, LMS, CRM, card swipe, smartphones of students, and surveys to delivery message nudges to students at opportune times”, where the “student event data sources” are databases of student records and the “incoming event data steams” include unstructured data, Para. [0169], “These multiple-data streams are converted into user-centric time series event data that adhere to defined entity-event taxonomies and stored as the user event logs 1502”, where “converted into” demonstrates the “streams” were not structured because they did not “adhere to defined entity-event taxonomies”; Fig. 3, where the “micro Intervention Delivery Subsystem 106” receives “Event Data”, which is provided over a computer network, see Fig. 15 and Para. [0168], “partner data 1508 encompassing various enterprise data from colleges and universities are ingested through an Application Programming Interface (API) 1510 that leverages third-party plugin tools 1512, especially for data sources managed through enterprise platform vendors' cloud services”, where and API “that leverages third-party plugin tools” and “cloud services” “to ingest data” from the smartphones “1520” of students at particular “colleges and universities” requires a computer network);
[converting] . . . the unstructured data [into structured data] . . . [by performing conversions] to classify the unstructured data . . . to classify the unstructured data into a plurality of classifications . . . (Para. [0169], “These multiple-data streams are converted into user-centric time series event data that adhere to defined entity-event taxonomies and stored as the user event logs 1502 so that open-source tools that target such data schema can be leveraged”, where the “conver[sion]” organizes the “data streams” into structured “entity-event taxonomies”; see also Para. [0167], “A user event log 1502 contains student event data stream (timestamped records of student activity)” and Fig. 14, where the column “Student info” includes classifications of data from the “data steams”, such as “ACT = 31” and “HS GPA = 3.8”);
. . . a plurality of unstructured data sources . . . the plurality of unstructured data sources . . . unstructured data source . . . unstructured data source . . . the plurality of unstructured data sources (Para. [0166], “one implementation of the micro intervention delivery subsystem 106 . . . uses incoming event data streams from multiple student event data sources, such as, but not limited to, SIS, LMS, CRM, card swipe, smartphones of students, and surveys to delivery message nudges to students at opportune times”, where the “student event data sources” are a plurality of sources, “multiple . . . data sources”, for unstructured data, “incoming event data steams”; see also Para. [0006], “A student data-to-insight-to-action-to-learning analytics system in accordance with an embodiment of the invention comprises memory and a processor, which is configured to compute student success predictions, student engagement predictions, and student impact predictions to interventions using at least linked-event features from multiple student event data sources”);
classifying, . . . , the unstructured data into the plurality of classifications (Para. [0167], “A user event log 1502 contains student event data stream (timestamped records of student activity)” and Fig. 14, where the column “Student info” includes a plurality of classifications of the unstructured data from the “data steams”, such as “ACT = 31” and “HS GPA = 3.8”);
based on the classifications of the unstructured data, identifying at least one of a plurality of points in the educational journey, the points including (i) points that are associated with hindering the particular student's progress in the educational journey and (ii) points that are associated with promoting the particular student's progress in the educational journey (Para. [0107], “the student impact prediction subsystem 104 builds models . . . using student information . . . as shown in FIG. 14”, where “the student impact prediction subsystem 104 builds models” based on the classifications of unstructured data, because the structured “entity-event taxonomies” are generated by classifying the unstructured “data streams”, see above; Para. [0036], “the student impact prediction subsystem 104 includes a multi-level linked-event feature extraction module 112”, where the “multi-level linked-event feature extraction module 112” “extract[s]” “linked-event feature[s]” to be “systematically analyzed”, Para. [0037], “The multi-level linked-event feature extraction module 112 provides the answer to the why question. For example, feature analysis shows new students with high ACT or SAT scores tend to persist at a lower rate when these students perform poorly on their mid-term exams. Furthermore, how they bounce back from such adversities can be a strong indicator of grit and future success. Such linked-event features can be systematically analyzed in terms of their predictive power, interpretability, engagement, and impact. FIG. 2 shows a table with examples of linked-event features divided into seven (7) categories in accordance with an embodiment of the invention”, where “features” like “high ACT” or “consistency” in “Academic performance” are points that are associated with promoting the particular student's progress in the educational journey and “features” like “Unmet need” for “financial aid” are points that are associated with hindering the particular student's progress in the educational journey);
using the points, generating prediction information of the particular student's progress in the educational journey (Para. [0039], “Using such real-time linked-event features coupled with background information, the multi-modal student success prediction module 114 next predicts student success in multiple dimensions, such as, but not limited to, academic success, persistence, switching majors, time to and credits at graduation, and post-graduation success”, where the “multi-modal student success prediction model 114” uses the points, “linked-event features”, to generate a prediction on “student success in multiple dimensions”, which, in combination with the predictions from the “Student Engagement Prediction Module 116” and the ”Student Impact Prediction Module 118 ”, form the prediction information of the particular student's progress in the educational journey, see Fig. 1 and Para. [0041], “Engagement and impact predictions made by the student engagement prediction module 116 and the student impact prediction module 118 complete the hierarchical three-level prediction cycle that connects predictive insights to actions to results”; for more information see Para. [0039] – [0042]),
the prediction information being consistent with a machine learning prediction model (Para. [0109] – [0112], “The student-engagement model in the student engagement prediction module 116 y.sub.E=f(x) has the following attributes: 3. Learning algorithm”, where the prediction information is consistent with a machine learning prediction model because the “prediction” is produced using a “model” with a “learning algorithm”; see also Para. [0041], “the student engagement prediction module 116 and the student impact prediction module 118 complete the hierarchical three-level prediction cycle that connects predictive insights to actions to results. These predictions require the analysis results of the impact analysis subsystem 108”, where the “predictions” are also consistent with a machine learning prediction model because it uses “results of the impact analysis system 108” to generate its predictions, and the prediction model can therefore be considered as comprising both components of “104” and “108”, where “108” in turn use machine learning prediction models, see Fig. 12 and Para. [0019], “FIG. 12 shows different learning algorithms that can be used by the tier-3 impact analysis module”);
generating a journey map that includes the prediction information (Para. [0005], “A student data-to-insight-to-action-to-learning analytics method in accordance with an embodiment of the invention comprises computing student success predictions, student engagement predictions, and student impact predictions to interventions ”, where the “predictions” are incorporated into a predictive mapping, see Fig. 14, of an educational journey of a student at an educational institution; see also Para. [0039], “By virtue of having a different subset of top predictors during various stages of a student's academic journey, higher-education (HE) institutions can develop more timely and context-aware student outreach programs and policies”)
and includes one or more of the points selected based on a magnitude of correlation between (i) the points and (ii) a predicted event indicated by the prediction information (Fig. 14, where the points include “ACT =31”, which is selected for, at least in part because “ACT or SAT scores” have a magnitude of correlation, “tend to . . . predictive power”, between the points, “high ACT”, and a predicted event indicated by the prediction information, “persist at a lower rate when these students perform poorly on their mid-term exams”, see Para. [0037], “The multi-level linked-event feature extraction module 112 provides the answer to the why question. For example, feature analysis shows new students with high ACT or SAT scores tend to persist at a lower rate when these students perform poorly on their mid-term exams . . . Such linked-event features can be systematically analyzed in terms of their predictive power, interpretability, engagement, and impact”),
wherein the selected points include points having a magnitude of correlation exceeding a predetermined threshold or include a top predetermined number of points by the magnitude of correlation (Para. [0037], “The multi-level linked-event feature extraction module 112 provides the answer to the why question. For example, feature analysis shows new students with high ACT or SAT scores tend to persist at a lower rate when these students perform poorly on their mid-term exams . . . Such linked-event features can be systematically analyzed in terms of their predictive power, interpretability, engagement, and impact” and Fig. 14, where, as discussed above, the one or more points are selected for based on a magnitude of correlation, which must exceed a threshold of being “reflect[ive of] only the best evidence-based practices” for “engagement and impact strategies”, see Para. [0045], “The evidence-based action knowledge database 102 works in concert with the lifecycle management subsystem 110 to ensure that engagement and impact strategies reflect only the best evidence-based practices over time as student characteristics and intervention strategies change over time”, and includes a top number of points, “selecting top features”, by magnitude of correlation, “answer to why students have such prediction scores can also be explained”, see Para. [0040], “The multi-modal student success prediction models generated by the multi-modal student success prediction module 114 . . . By virtue of competing and selecting top features for various models built for different student segments, the answer to why students have such prediction scores can also be explained”; see also Para. [0004], “The evidence-based action knowledge database is updated by executing a multi-tier impact analysis on impact results of applied interventions”, where both the top features and threshold are predetermined by “update[ing]” the “evidence-based action knowledge database” through “executing a multi-tier impact analysis on impact results of applied interventions” prior to deployment);
wherein generating the journey map comprises including, in the journey map, extracted portions of the unstructured data (Para. [0167], “A user event log 1502 contains student event data stream (timestamped records of student activity)” and Fig. 14, where the column “Student info” of the journey map, Fig. 14, includes a plurality of classifications of extracted portions of the unstructured data from the “data steams”, such as “computer science major” and “HS GPA = 3.8”)
that are extracted by identifying phrases used to categorize the unstructured data into the points (Fig. 4, where extraction occurs by identifying phrases with “Engagement rules”, such as “Major = accounting” or “Course = English”, which are used to extract features and to categorize the unstructured data into the points, see Para. [0042], “Engagement rules are expressed in terms of linked-event features and prediction scores to isolate opportune moments for reaching out to students”; see also Para. [0167], “The incoming event data stream consists of passive sensing data with student opt-in and institutional data . . . A user event log 1502 contains student event data stream (timestamped records of student activity). A nudge log 1504 contains triggered nudges or messages to be delivered to particular students at specific times based on engagement rules being fired”; see generally Fig. 14); and
transmitting the prediction information and the journey map over the computer network to an administrator machine (Para. [0036], “In some embodiments, at least some of these components of the student impact prediction subsystem 104 are implemented as one or more software programs running in one or more computer systems using one or more processors and memories associated with the computer systems. These components may reside in a single computer system or distributed among multiple computer systems, which may support cloud computing”, where embodiments of “the student impact prediction subsystem 104”, which include “114”, “116”, and “118”, see Fig. 1, are “cloud” “distributed” as “multiple computer systems”, which is within the broadest reasonable interpretation of a computer network; see also Para. [0021], “FIG. 14 shows representative data samples from the tier-1 impact analysis module that can be used to build student engagement and impact models” and Para. [0059, “the three-tier impact analysis sub-system 108 includes a tier-1 impact analysis module 120, a tier-2 impact analysis module 122, a tier-3 impact analysis module 124 and an impact result packing module 126 . . . These components may reside in a single computer system or distributed among multiple computer systems, which may support cloud computing”, where the journey map, as depicted in “FIG. 14”, is transmitted over a computer network, “distributed among multiple computer systems, which may support cloud computing”, to “be used to build student engagement and impact models”, where the machine associated with this functionality can reasonably be described as an administrator machine because it administers the “building” and thus controls the “models”, see Fig. 16, Fig. 18, and Para. [0176], “FIG. 16 depicts a homepage that illustrate how such connected, predictive, and action insights can be communicated to various stakeholders to create a virtuous circle in accordance with an embodiment of the invention”, where “predictive” information is provided to homepages and dashboards of administrative “stakeholders” which must have associated machines connected to the “cloud” “distributed” “computer systems”; see generally Para. [0046], “A student in a pilot program can receive treatment or micro intervention defined as contact between a student and an institutional entity encompassing, but not limited to, faculty, advisors, administrators, student mentors/mentees, and personal digital Sherpas or guides”).
Kil does not explicitly disclose . . . normalizing . . . consistent with a machine learning classification model . . . for input into the machine learning classification model . . . wherein training the machine learning classification model comprises training the machine learning classification model using source-separated training data sets respectively corresponding to. . . , including maintaining separation of . . . during the training such that points unique to a smaller-volume . . . are surfaced and not washed away by a larger-volume . . . , including (i) maintaining the source-separated training data sets as separate training data sets corresponding to . . . and (ii) training using the source-separated training data sets without combining the source-separated training data sets into a single training data set . . . using the machine learning classification model . . .
However, Kumar teaches . . . normalizing [data] . . . consistent with a machine learning classification model [to classify the data] . . . for input into the machine learning classification model . . . (Para. [0087], “each feature in some or all of the features (e.g., as illustrated in TABLE 1 above) may be normalized to a range of values between zero (0) and one (1) and sent/provided to a keyword classifier implemented by (or operating in conjunction with) natural language processor 208”; Para. [0088], “natural language processor 208 builds or implements a classification model (e.g., a random forest classification model, etc.) with the keyword classifier. The classification model may be used by the keyword classifier to predict the probability (“p”) of the filtered token being a keyword for the knowledge domain”)
wherein training the machine learning classification model comprises training the machine learning classification model . . . (Para. [0088], “In some embodiments, natural language processor 208 builds or implements a classification model (e.g., a random forest classification model, etc.) with the keyword classifier. The classification model may be used by the keyword classifier to predict the probability (“p”) of the filtered token being a keyword for the knowledge domain”, where a person of ordinary skill in the art would understand a “random forest classification model” to be trained with training data, see Para. [004], “Machine learning may be implemented based on a set of training data to train potentially complex models and algorithms for making predictions and further based on a set of test data to measure accuracies and robustness in the predictions made with the complex models and algorithms as trained with the set of training data”)
using the machine learning classification model . . . (The classification model may be used by the keyword classifier to predict the probability (“p”) of the filtered token being a keyword for the knowledge domain”)
Before the effective filing date of the invention, it would have been obvious to one of ordinary skill in the art to combine the converting of unstructured data by performing conversions to classify the instructed data into a plurality of classifications of Kil with the normalization of data for use by a machine learning classification model to classify the data, wherein the data is input into the classification model for model use and the model is trained of Kumar in order to represent features on a comparable scale, which will produce more dependable results (Kumar, Para. [0087], “each feature in some or all of the features (e.g., as illustrated in TABLE 1 above) may be normalized to a range of values between zero (0) and one (1) and sent/provided to a keyword classifier implemented by (or operating in conjunction with) natural language processor 208”; Kumar, Pg. 6, Col. 2, Table 1, where use of cross “Category” “Features”, such as features across categories of “Frequency”, “Structure Type”, and “Relationship” may produce skewed results if not first “normalized to a range of values”), and to utilize machine learning techniques to complete a wide variety of complex classification tasks (Kumar, Para. [0004], “Machine learning may be implemented based on a set of training data to train potentially complex models and algorithms for making predictions and further based on a set of test data to measure accuracies and robustness in the predictions made with the complex models and algorithms as trained with the set of training data”; see also Kumar, Para. [0088], “natural language processor 208 builds or implements a classification model (e.g., a random forest classification model, etc.) with the keyword classifier. The classification model may be used by the keyword classifier to predict the probability (“p”) of the filtered token being a keyword for the knowledge domain”, where the NLP model can effectively filter token based on learned domain knowledge), where both the normalization and use of an NLP ML classifier contribute to a system with maximal use of available data (see Kumar, Para. [0028], “Techniques as described herein can be used to enable AI models/algorithms operating with other applications and other use cases to make maximal/optimal/efficient use of the historic data available”), which will have increased prediction accuracy through training with the plurality of unstructured data sources (Kumar, Para. [0004], “Machine learning may be implemented based on a set of training data to train potentially complex models and algorithms for making predictions and further based on a set of test data to measure accuracies and robustness in the predictions made with the complex models and algorithms as trained with the set of training data”).
Additionally, Sajjad teaches . . . using source-separated training data sets respectively corresponding to . . . [a plurality of data sets], including maintaining separation of . . . [the plurality of data sets] during the training such that points unique to a smaller-volume . . . are surfaced and not washed away by a larger-volume . . . , including (i) maintaining the source-separated training data sets as separate training data sets corresponding to . . . [a plurality of data sets] and (ii) training using the source-separated training data sets without combining the source-separated training data sets into a single training data set . . . (Pg. 2, Col. 1, Fig. 1 and Pg. 2, Col. 2, Para. 2, “we propose a stacking method that uses multi-level fine-tuning to train a system. Figure 1 (second row) shows the complete procedure: first, the model is trained on the out-of-domain data Do1 for N epochs; training is resumed from N + 1-th epoch to the M-th epoch but using the next available out-of domain data Do2 ; repeat the process till all of the available out-of-domain corpora have been used; in the last step, resume training on the in-domain data Di for a few epochs. The resulting model has seen all of the available data as in the case of the data concatenation approach. However, here the system learns from the data domain by domain. We call this technique model stacking”, where a plurality of data sets corresponding to respective “domain data[sets]” are used for “model . . . train[ing]”, and where the separation is maintained by using specific “domain data” at each “epoch” without combining the training data into a single training data set, “here the system learns from the data domain by domain”, such that points unique to a smaller-volume data set are surfaced and not washed away by a larger-volume dataset, see Pg. 5, Col. 1, Para. 1, “The best combination for the Arabic-English language pair started with the UN data, fine-tuned on OPUS and then fine-tuned on TED” and Pg. 3, Col. 2, Table 1, where the smaller volume data set, “TED 229k 3.7M 4.7M” is surfaced by placement last in the “fine-tun[ing]” which prevents it from being washed away by a larger volume dataset, “UN 18.3M 433M 494M” or “OPUS 22.4M 139M 195M”) (Examiner’s note – the recitation of “such that points unique to a smaller-volume . . . are surfaced and not washed away by a larger-volume” raises questions as to its limiting effect, see MPEP 2103, “The following types of claim language may raise a question as to its limiting effect: . . . (A) statements of intended use or field of use”).
Before the effective filing date of the invention, it would have been obvious to one of ordinary skill in the art to combine the training of the machine learning classification model using a plurality of unstructured data sources of Kil in view of Kumar with the use of source-separated training data sets respectively corresponding to a plurality of data sets, where separation of the plurality of data sets is maintained during training without combining the source-separated training data sets into a single training data set, such that such points unique to a smaller-volume are surfaced and not washed away by a larger-volume of Sajjad in order to achieve superior model performance (Sajjad, Pg. 5, Col. 1, Para. 1, “The best combination for the Arabic-English language pair started with the UN data, fine-tuned on OPUS and then fine-tuned on TED” and Sajjad, Pg. 1, Abstract, “Our findings show that the best translation quality can be achieved by building an initial system on a concatenation of available out-of-domain data and then fine tuning it on in-domain data”), using a flexible system where specific data sets, such as those with lower volume, can be prioritized (Sajjad, Pg. 2, Col. 2, Para. 3, “The model stacking and fine-tuning approaches have the advantage of seeing the in-domain data in the end of training, thus making the system parameters more optimized for the in-domain data. They also provide flexibility in extending an existing model to any new domain without having to retrain the complete system again on the available corpora”).
Regarding Claim 2, Kil in view of Kumar and Sajjad teach the method of claim 1, wherein the prediction information is included in a journey map comprising an indication of a plurality of points, including summary information for the plurality of points (Kil, Fig. 14, where the “Student Information” column includes an indication of a plurality of points, such as “student loan” and “proprietary school”, and the “Engagement rule” includes summary information for the points, such as “high loan amount”; see generally Kil, Para. [0039] - [0042], “Using such real-time linked-event features coupled with background information, the multi-modal student success prediction module 114 next predicts student success . . . Engagement rules are expressed in terms of linked-event features and prediction scores to isolate opportune moments for reaching out to students”).
Regarding Claim 3, Kil in view of Kumar and Sajjad teach the method of claim 1, wherein the unstructured data comprises at least one of mentor notes, email interactions, assessment responses, instructor notes, social media posts, course surveys, personality test responses, or aptitude test responses (Kil, Para. [0167], “The incoming event data stream consists of passive sensing data with student opt-in and institutional data consisting of, but not limited to, SIS, LMS, CRM, card swipe data, and location beacon data”, where, as discussed above, the “event data stream” includes the unstructured data, see Kil, Para. [0005], “A student data-to-insight-to-action-to-learning analytics method in accordance with an embodiment of the invention comprises . . . using at least linked-event features from multiple student event data sources”, which must include at least include personality test responses in order for “Survey: . . . RNL CSI” to be “from multiple student event data sources”, see Kil, Fig. 2 and Kil, Para. [0009], “FIG. 2 shows a table with examples of linked-event features divided into seven (7) categories in accordance with an embodiment of the invention”).
Regarding Claim 4, Kil in view of Kumar and Sajjad teach the method of claim 1, further comprising: receiving over the computer network, structured data (Kil, Para. [0036], “In some embodiments, at least some of these components of the student impact prediction subsystem 104 are implemented as one or more software programs running in one or more computer systems using one or more processors and memories associated with the computer systems. These components may reside in a single computer system or distributed among multiple computer systems, which may support cloud computing”, where embodiments of “the student impact prediction subsystem 104”, which include “114”, “116”, and “118”, see Kil, Fig. 1, are “cloud” “distributed” as “multiple computer systems”, which is within the broadest reasonable interpretation of a computer network, and therefore the transmission from the “Evidence-Based Action Knowledge Database 102” to the components of the “Student Impact Prediction Subsystem 104” is a receiving over the computer network, and the transmission is of the structured data, see Kill, Fig. 14, where the column “Student info” includes classifications of data from the “data steams”, such as “ACT = 31” and “HS GPA = 3.8”; Kil, Para. [0107], “the student impact prediction subsystem 104 builds models . . . using student information . . . as shown in FIG. 14”, where “the student impact prediction subsystem 104 builds models” uses the structured “entity-event taxonomies” from “102”, see Kil, Fig. 1 and Kil, Para. [0091], “the evidence-based action knowledge database (EAKD) 102 can be used to build and deploy the student engagement and impact prediction models”);
and wherein the structured data is used in the machine learning prediction model in generating the prediction information of the particular student's progress in the educational journey (Kill, Para. [0036], “the student impact prediction subsystem 104 includes a multi-level linked-event feature extraction module 112”, where the “multi-level linked-event feature extraction module 112” “extract[s]” “linked-event feature[s]” from the structured data to be “systematically analyzed”, Kil, Para. [0037], “Such linked-event features can be systematically analyzed in terms of their predictive power, interpretability, engagement, and impact. FIG. 2 shows a table with examples of linked-event features divided into seven (7) categories in accordance with an embodiment of the invention”; Kil, Para. [0039], “Using such real-time linked-event features coupled with background information, the multi-modal student success prediction module 114 next predicts student success in multiple dimensions, such as, but not limited to, academic success, persistence, switching majors, time to and credits at graduation, and post-graduation success”, where the “multi-modal student success prediction model 114” uses the friction and achievement points, “linked-event features”, to generate a prediction on “student success in multiple dimensions”, which, in combination with the predictions from the “Student Engagement Prediction Module 116” and the ”Student Impact Prediction Module 118 ”, form the prediction information of the particular student's progress in the educational journey, see Kil, Fig. 1 and Kil, Para. [0041], “Engagement and impact predictions made by the student engagement prediction module 116 and the student impact prediction module 118 complete the hierarchical three-level prediction cycle that connects predictive insights to actions to results”; for more information see Kil, Para. [0039] – [0042]).
Regarding Claim 5, Kil in view of Kumar and Sajjad teach the method of claim 4, wherein the structured data and unstructured data comprises at least one of mentor notes, email interactions, helpdesk tickets, program information, assigned grades, discipline write ups, assessment responses, financial aid status, transferred credits, academic resource interactions, instructor notes, marketing data, social media posts, governmental body reports, course survey responses, personality test responses, or aptitude test responses (Kil, Fig. 14, where the structured data includes relational information such as at least financial aid status, “student loan = $500/credit” and “Pell grant recipient”, and, as a result, the unstructured data used to generate the structured data, see Kil, Para. [0169], “These multiple-data streams are converted into user-centric time series event data that adhere to defined entity-event taxonomies and stored as the user event logs 1502 so that open-source tools that target such data schema can be leveraged”, must have also comprised this information).
Regarding Claim 6, Kill in view of Kumar and Sajjad teach the method of claim 4, wherein the structured data is used to generate points (Kill, Fig. 14, where the column “Student info” includes classifications of data from the “data steams”, such as “ACT = 31” and “HS GPA = 3.8”; Kil, Para. [0107], “the student impact prediction subsystem 104 builds models . . . using student information . . . as shown in FIG. 14”, where “the student impact prediction subsystem 104 builds models” uses the structured “entity-event taxonomies” generated by classifying the unstructured “data streams”; Kil, Para. [0036], “the student impact prediction subsystem 104 includes a multi-level linked-event feature extraction module 112”, where the “multi-level linked-event feature extraction module 112” “extract[s]” “linked-event feature[s]” to be “systematically analyzed”, Kil, Para. [0037], “The multi-level linked-event feature extraction module 112 provides the answer to the why question. For example, feature analysis shows new students with high ACT or SAT scores tend to persist at a lower rate when these students perform poorly on their mid-term exams. Furthermore, how they bounce back from such adversities can be a strong indicator of grit and future success. Such linked-event features can be systematically analyzed in terms of their predictive power, interpretability, engagement, and impact. FIG. 2 shows a table with examples of linked-event features divided into seven (7) categories in accordance with an embodiment of the invention”, where “features” like “high ACT” or “consistency” in “Academic performance” are points that promote progress and “features” like “Unmet need” for “financial aid” are points that hinder progress).
Regarding Claim 7, Kill in view of Kumar and Sajjad teach the method of claim 4, wherein the structured data is used to refine the machine learning prediction model (Kil, Para. [0114], “The evidence-based action knowledge database (EAKD) 102 stores tier-1, tier-2, and tier-3 impact results to promote the development and retraining of student-engagement and student-impact prediction models”, where the “impact results”, which are the basis for “development and retraining” are in turn based on the outputs of “The micro intervention delivery subsystem 106”, see Kil, Para. [0043] – [0044], “the micro intervention delivery subsystem 106 operates to deliver micro interventions . . . the three-tier impact analysis subsystem 108 operates to look for results of delivered micro interventions in several time scales using three-tier analyses”, which, uses the structured data to determine the output from “The micro intervention delivery subsystem 106”, see Kil, Para. [0166] – [0169], “one implementation of the micro intervention delivery subsystem 106 in accordance with an embodiment of the invention is shown as a nudge delivery subsystem 1500. The nudge delivery subsystem 1500 uses incoming event data streams from multiple student event data sources . . . A nudge log 1504 contains triggered nudges or messages to be delivered to particular students at specific times based on engagement rules being fired . . . These multiple-data streams are converted into user-centric time series event data that adhere to defined entity-event taxonomies and stored as the user event logs 1502 so that open-source tools that target such data schema can be leveraged”, therefore, the refining, at least indirectly, uses the structured data based on the classifications; see also Kil, Para. [0073] – [0074] and [0077] – [0078] for additional information on improving the impact analysis system).
Regarding Claim 8, Kil in view of Kumar and Sajjad teach the method of claim 1, wherein the unstructured data, based on the classifications, is used to refine the machine learning prediction model (Kil, Para. [0114], “The evidence-based action knowledge database (EAKD) 102 stores tier-1, tier-2, and tier-3 impact results to promote the development and retraining of student-engagement and student-impact prediction models”, where the “impact results”, which are the basis for “development and retraining” are in turn based on the outputs of “The micro intervention delivery subsystem 106”, see Kil, Para. [0043] – [0044], “the micro intervention delivery subsystem 106 operates to deliver micro interventions . . . the three-tier impact analysis subsystem 108 operates to look for results of delivered micro interventions in several time scales using three-tier analyses”, which, based on the classifications, uses the unstructured data to create structured data, which is used to determine the output from “The micro intervention delivery subsystem 106”, see Kil, Para. [0166] – [0169], “one implementation of the micro intervention delivery subsystem 106 in accordance with an embodiment of the invention is shown as a nudge delivery subsystem 1500. The nudge delivery subsystem 1500 uses incoming event data streams from multiple student event data sources . . . A nudge log 1504 contains triggered nudges or messages to be delivered to particular students at specific times based on engagement rules being fired . . . These multiple-data streams are converted into user-centric time series event data that adhere to defined entity-event taxonomies and stored as the user event logs 1502 so that open-source tools that target such data schema can be leveraged”, therefore, the refining, at least indirectly, uses the unstructured data based on the classifications; see also Kil, Para. [0073] – [0074] and [0077] – [0078] for additional information on improving the impact analysis system).
Regarding Claim 9, Kil in view of Kumar and Sajjad teach the method of claim 1, wherein the machine learning classification model comprises a natural language processing model (Kumar, Para. [0087], “each feature in some or all of the features (e.g., as illustrated in TABLE 1 above) may be . . . sent/provided to a keyword classifier implemented by . . . natural language processor 208”; Kumar, Para. [0088], “natural language processor 208 builds or implements a classification model (e.g., a random forest classification model, etc.) with the keyword classifier. The classification model may be used by the keyword classifier to predict the probability (“p”) of the filtered token being a keyword for the knowledge domain”).
The reasons for obviousness were provided in regard to the rejection of claim 1 above, and remain applicable here.
Regarding Claim 10, Kil in view of Kumar and Sajjad teach the method of claim 1, wherein the machine learning prediction model comprises a root cause analysis model (Kil, Para. [0109] – [0112], “The student-engagement model in the student engagement prediction module 116 y.sub.E=f(x) has the following attributes: 3. Learning algorithm”; Kil, Para. [0041], “the student engagement prediction module 116 and the student impact prediction module 118 complete the hierarchical three-level prediction cycle that connects predictive insights to actions to results. These predictions require the analysis results of the impact analysis subsystem 108”, where the machine learning model of “116” can reasonably be considered a root cause analysis model because it analyzes the outputs of “108” to generate its outputs, “116 . . . connects predictive insights to actions to results”, which “require[s] the analysis results of the impact analysis subsystem 108”, which “108” outputs are information on root cause “intervention” for “results”, see Kil, Para. [0044], “The tier-3 impact analysis measures the results of students exposed to various micro interventions using term-level metrics, such as, but not limited to, semester grade point average (GPA), successful course completion, engagement, persistence, graduation, job placement, and salary” and Kil, Para. [0179], “The analytics system 100 also provides three-tier impact analysis that resolves results-attribution ambiguity through micro-pathway construction between actions and results, which serves as an engine to both engagement and impact predictions”, where “micro-pathway construction between actions and results” provides root cause information for 116 and 118 “engagement and impact predictions”).
Regarding Claim 11, Kil in view of Kumar and Sajjad teach the method of claim 1, further comprising: using the machine learning prediction model to generate intervention output suggesting action to alter the generated prediction information (Kil, Para. [0041] – [0042], “Engagement and impact predictions made by the student engagement prediction module 116 and the student impact prediction module 118 . . . connects predictive insights to actions to results . . . with a particular emphasis on parameterization of intervention, student, and prediction characteristics . . . Engagement rules are expressed in terms of linked-event features and prediction scores to isolate opportune moments for reaching out to students”, where “116”, which contains the machine learning prediction model, is used to develop “Engagement rules” based on “intervention, student, and [engagement and impact] prediction characteristics”, which demonstrates the rules are based on the level of impact at altering the student success component of the generated information; Kil, Para. [0056], “The micro intervention delivery sub-system 106 operates to systematically evaluate a number of engagement rules . . . The micro intervention delivery subsystem 106 then facilitates delivery of an appropriate micro intervention corresponding to the highest ranked engagement rule”, where the “engagement rule[s]” used to generate the “appropriate micro intervention”);
and transmitting the intervention output over the computer network to the administrator machine (Kil, Para. [0043], “The micro intervention delivery subsystem 106 operates to deliver micro interventions”, where “106” “deliver[s] micro interventions” that are accessible to “administrators”, see generally Kil, Para. [0046], “Treatment or micro intervention: A student in a pilot program can receive treatment or micro intervention defined as contact between a student and an institutional entity encompassing, but not limited to, faculty, advisors, administrators, student mentors/mentees, and personal digital Sherpas or guides”; Kil, Fig. 1 and Kil, Para. [0059], “the three-tier impact analysis sub-system 108 . . . components may reside in a single computer system or distributed among multiple computer systems, which may support cloud computing”, where “106” delivers transmitted interventions to “cloud” “distributed” “computer systems”, which are within the broadest reasonable interpretation of over a computer network; Fig. 16-18 and Para. [0023] – [0025], “[0023] FIG. 16 depicts a homepage that illustrate how connected, predictive, and action insights can be communicated to various stakeholders to create a virtuous circle in accordance with an embodiment of the invention. FIG. 17 depicts a drill-down initiative page in accordance with an embodiment of the invention. FIG. 18 shows an example of a real-time student success program impact dashboard that can be provided by the student data-to-insight-to-action-to-learning analytics system”, where, though Fig. 1 is focused on the network transmission to “108”, network transmission to administrative “stakeholder” machines is required to allow for “dashboard” “initiative” “page[s]” to be displayed on “stakeholder” machines).
Regarding Claim 12, Kil in view of Kumar and Sajjad teach a computing system comprising: one or more processors; and one or more computer-readable media having stored thereon instructions that are executable by the one or more processors (Kil, Para. [0004] - [0005], “Student data-to-insight-to-action-to-learning analytics system and method use an evidence-based action knowledge database to compute student success predictions, student engagement predictions, and student impact predictions to interventions . . . In some embodiments, the steps of this method are performed when program instructions contained in a computer-readable storage medium are executed by one or more processors”);
network hardware configured to . . . (Kil, Para. [0166], “one implementation of the micro intervention delivery subsystem 106 . . . uses incoming event data streams from multiple student event data sources, such as, but not limited to, SIS, LMS, CRM, card swipe, smartphones of students, and surveys to delivery message nudges to students at opportune times” and Kil, Para. [0168], “partner data 1508 encompassing various enterprise data from colleges and universities are ingested through an Application Programming Interface (API) 1510 that leverages third-party plugin tools 1512, especially for data sources managed through enterprise platform vendors' cloud services”, where the network for “data stream” “delivery” requires hardware to be functional, such as “smartphones”)
a categorization engine comprising a machine learning classification model, implemented by the one or more processors and the instructions, configured to . . . (Kil, Fig. 1, where, as discussed in detail above, the “Micro Intervention Delivery Subsystem 106” performs classification of unstructured data, which is a categorization engine implemented by one or more processors, see Kil, Para. [0004] - [0005], “Student data-to-insight-to-action-to-learning analytics system and method use an evidence-based action knowledge database to compute student success predictions, student engagement predictions, and student impact predictions to interventions . . . In some embodiments, the steps of this method are performed when program instructions contained in a computer-readable storage medium are executed by one or more processors”, and which, in view of Kumar, comprises an ML classification model, see Kumar, Para. [0088], “natural language processor 208 builds or implements a classification model (e.g., a random forest classification model, etc.) with the keyword classifier. The classification model may be used by the keyword classifier to predict the probability (“p”) of the filtered token being a keyword for the knowledge domain”)
a prediction engine comprising a machine learning prediction model, implemented by the one or more processors and the instructions, configured to (Kil, Fig. 1, where, as discussed in detail above, the “Student Impact Prediction Subsystem 104” and “Three-tier Impact Analysis Subsystem 108” function to generate predictions using a machine learning prediction model, and are thus, collectively a prediction engine, which can be implemented by one or more processors, see Kil, Para. [0004] - [0005], “Student data-to-insight-to-action-to-learning analytics system and method use an evidence-based action knowledge database to compute student success predictions, student engagement predictions, and student impact predictions to interventions . . . In some embodiments, the steps of this method are performed when program instructions contained in a computer-readable storage medium are executed by one or more processors”; see also Kil, Para. [0109] – [0112], “The student-engagement model in the student engagement prediction module 116 y.sub.E=f(x) has the following attributes: 3. Learning algorithm”) . . .
and wherein the network hardware is configured to . . . (Kil, Para. [0036], “In some embodiments, at least some of these components of the student impact prediction subsystem 104 are implemented as one or more software programs running in one or more computer systems using one or more processors and memories associated with the computer systems. These components may reside in a single computer system or distributed among multiple computer systems, which may support cloud computing”; Kil, Fig. 16 and Kil, Fig. 18, where the “computing systems” include hardware to provide the “predictions” to “stakeholder” “dashboards”, see Kil, Para. [0176], “FIG. 16 depicts a homepage that illustrate how such connected, predictive, and action insights can be communicated to various stakeholders to create a virtuous circle in accordance with an embodiment of the invention”)
The reasons for obviousness were provided in regard to the rejection of Claim 1 above, and remain applicable here. Also, the remaining limitations are substantially the same as limitations of Claim 1, therefore it is rejected under the same rationale.
Regarding Claim 13, the additional elements of the dependent claim are substantially the same as limitations of Claim 2, therefore it is rejected under the same rationale.
Regarding Claim 14, the additional elements of the dependent claim are substantially the same as limitations of Claim 3, therefore it is rejected under the same rationale.
Regarding Claim 15, the additional elements of the dependent claim are substantially the same as limitations of Claim 8, therefore it is rejected under the same rationale.
Regarding Claim 16, the additional elements of the dependent claim are substantially the same as limitations of Claim 9, therefore it is rejected under the same rationale.
Regarding Claim 17, the additional elements of the dependent claim are substantially the same as limitations of Claim 10, therefore it is rejected under the same rationale.
Regarding Claim 18, the additional elements of the dependent claim are substantially the same as limitations of Claim 11, therefore it is rejected under the same rationale.
Regarding Claim 19, Kil in view of Kumar and Sajjad teach a computing system comprising one or more processors (Kil, Abstract, “Student data-to-insight-to-action-to-learning analytics system”; Kil, Para. [0006], “A student data-to-insight-to-action-to-learning analytics system in accordance with an embodiment of the invention comprises memory and a processor”);
and one or more computer-readable media having stored thereon instructions that are executable by the one or more processors to configure the computer system to perform predictive journey mapping, including instructions that are executable to configure the computer system to perform at least the following: . . . (Kil, Para. [0005], “In some embodiments, the steps of this method are performed when program instructions contained in a computer-readable storage medium are executed by one or more processors”; Kil, Abstract, “Student data-to-insight-to-action-to-learning analytics system and method use an evidence-based action knowledge database to compute student success predictions, student engagement predictions, and student impact predictions to interventions”).
The reasons for obviousness were provided in regard to the rejection of Claim 1 above, and remain applicable here. The remaining limitations are substantially the same as limitations of Claim 1, therefore it is rejected under the same rationale.
Regarding Claim 20, Kil in view of Kumar and Sajjad teach the computing system of claim 19, wherein the prediction information is included in a journey map comprising an indication of a plurality of the friction points or a plurality of the achievement points, including summary information for the plurality of the friction points or the plurality of the achievement points (Kil, Fig. 14, where the “Student Information” column includes an indication of a plurality of friction or achievement points, such as “student loan” and “proprietary school”, and the “Engagement rule” includes summary information for the points, such as “high loan amount”; see generally Kil, Para. [0039] - [0042], “Using such real-time linked-event features coupled with background information, the multi-modal student success prediction module 114 next predicts student success . . . Engagement rules are expressed in terms of linked-event features and prediction scores to isolate opportune moments for reaching out to students”).
Response to Amendment
Applicant's arguments filed on March 27th, 2026 have been fully considered. Each argument is addressed in detail below.
I. Applicant argues the objections to the specification should be withdrawn (Applicant’s Remarks, 03/27/2026, Pg. 15, Section “Specification Objection”).
Applicant’s amendments to the specification have overcome each and every objection to the specification, as previously set forth in the October 27th, 2025 Office Action. As a result, these objections have been withdrawn.
II. Applicant argues the objections to the claims should be withdrawn (Applicant’s Remarks, 03/27/2026, Pg. 15, Section “Claim Objections”).
Applicant’s amendments to the claims have overcome each and every objection to the claims, as previously set forth in the October 27th, 2025 Office Action. As a result, these objections have been withdrawn.
III. Applicant argues the rejections of the claims, under 35 USC § 112, should be withdrawn (Applicant’s Remarks, 03/27/2026, Pg. 15, Section “Section 35 U.S.C. § 112 Rejections”).
Applicant’s amendments to the claims have overcome some, but not all, of the rejections of the claims under 35 USC § 112, as previously set forth in the October 27th, 2025 Office Action. While some rejections have been withdrawn, the remaining issues are discussed in detail above. For clarity of the record, it is worth point out that applicant’s amendments intended to correct indefiniteness related to the use of the relative term “journey” were deemed sufficient with regard to the recitations of the “journey map” element, but insufficient with regard to the recitations of the “educational journey” element (see, for example, Claim 1).
Additionally, as discussed in detail above, Applicant’s amendments introduce new indefiniteness, which necessitate new grounds for rejection under 35 USC § 112.
IV. Applicant argues the rejections of the claims, under 35 USC § 101, should be withdrawn (Applicant’s Remarks, 03/27/2026, Pg. 15-18, Section “Section 35 U.S.C. § 101 Rejections”).
1) First, Applicant argues the claims “recite[] a specific improvement in Al technology: a training-data regimen that changes how the machine-learning classification model is trained and how it behaves in the presence of heterogeneous, multi-source unstructured datasets, together with concrete machine outputs (a journey map including prediction information, correlation-selected points, and extracted portions of unstructured data)”, which “is precisely the category of claim recognized as eligible under the Appeals Review Panel's precedential decision in Ex parte Desjardins: a claim that may include mathematical concepts inherent in machine learning, but that is not directed to an abstract idea because it recites a concrete mechanism improving the Al system itself” (Pg. 16, Para. 1-2).
Additionally, Applicant asserts the catastrophic forgetting of Desjardins is analogous to the smaller source washout of the instant claims because “training using source-separated training data sets while maintaining separation during training (including expressly not combining the sets into a single training set) such that points unique to a smaller-volume unstructured data source are surfaced and not washed away by a larger- volume unstructured data source”, which “is not a result-only statement”, but instead “is tied to a specific training constraint ("source-separated" + "do not combine") and a specific technical effect (preserving/surfacing minority-source patterns)” (Pg. 17, Para. 2-6).
Here, the asserted improvements related to training of a machine learning classification model are recited in the specification and reflected in the claims in a merely conclusory manner without the detail necessary to be apparent to a person of ordinary skill in the art (see MPEP 2106.04(d)(1)). Specifically, the claims as currently formulated merely asserts an improvement to model training will result from source-separated data that is not combined, which in turn is recited broadly such as to amount to merely claiming the idea of a solution (see MPEP 2106.05(a)) which has broad applicability across many fields of endeavor (see MPEP 2106.05(f)). As a result, these bare assertions fail to transform the claim into patent eligible subject matter because there are no meaningful limitations associated with the purported improvements (see MPEP 2106.04(d)(1)). Furthermore, in relation to the mental processes of prediction, generation, correlation, and extraction, the recitations of machine learning model outputs amount to a drafting effort to monopolize abstract ideas (see MPEP 2106.04(d)). This meaningfully distinguished the claims from the subject matter at issue in the Desjardins decision, as well as the related guidance disseminated on the subject.
As a result, the arguments are not persuasive.
2) Second, Applicant argues the amended claims are not mental processes because it “requires a machine-learning classification model trained using source- separated training data sets that correspond to a plurality of unstructured data sources, with separation maintained in operative terms-specifically, (i) maintaining the source-separated training data sets as separate training data sets and (ii) training using the source-separated training data sets without combining them into a single training data set. This is a concrete, machine-implemented training architecture; it is not performable in the human mind, nor is it a mere instruction to use an ML model” (Pg. 16, Para. 3-4) (internal quotation marks omitted).
However, a claim that requires a computer may still recite a mental process (see MPEP 2106.04(a)(2)).
As a result, the argument is not persuasive.
3) Third, Applicant argues the amended claims are integrated into a practical application because it recites “a concrete, machine-implemented training architecture” that is not “mere instruction to use an ML model” and it “requires generating a journey map that includes: (i) prediction information, (ii) one or more points selected based on a computed magnitude of correlation between points and a predicted event, with inclusion defined by threshold/top-N, and (iii) extracted portions of the unstructured data, extracted by identifying phrases used to categorize the unstructured data into the points. These limitations define an end-to-end computerized pipeline producing a concrete output artifact (a journey map populated with extracted textual portions and correlation-selected points) and transmitting it to an administrator machine. The claim therefore is integrated into a practical application and is not properly characterized as insignificant extra-solution activity" (Pg. 16-17, Para. 3-1) (internal quotation marks omitted).
Here, as discussed in detail above, the recitations relating to generation of a journey map amount to mental processes. As a result, the question is whether the additional elements demonstrate that the claim as a whole integrates the judicial exception into a practical application (see MPEP 2106.04(d)). As also discussed in detail above, the recitations of computer components relating to machine learning, data transmission, and configuration of hardware are properly characterized as mere instructions to apply the exception on generic computer hardware because the components are recited at a high level of generality and as merely a tool to perform the process (see MPEP 2106.05(f)). Whereas the transmission of data itself is properly characterized as insignificant extra-solution activity that is well-understood, routine, and conventional because it amounts to receiving or transmitting data over a network (see MPEP 2106.05(g) and MPEP 2106.05(d)).
As a result, the argument is not persuasive.
4) Fourth, Applicant argues “the amended claim now expressly recites the training-data regimen details the Examiner requested in interview (source-separated training data sets maintained as separate; not combined), and expressly recites the technical effect (surfacing smaller-source points not washed away). This is an operational constraint on the training pipeline that changes model behavior-exactly the type of improvement that Desjardins instructs should be credited at §101” (Pg. 18, Para. 1).
Here, Applicant is correct that the Examiner recommended incorporating additional amendments relating to the source-separated training data. However, as discussed in detail above, the additional amendments are insufficient to render the claims patent eligible because the asserted improvements related to training of a machine learning classification model are recited in the specification and reflected in the claims in a merely conclusory manner without the detail necessary to be apparent to a person of ordinary skill in the art (see MPEP 2106.04(d)(1)). As a result, these bare assertions fail to transform the claim into patent eligible subject matter because there are no meaningful limitations associated with the purported improvements (see MPEP 2106.04(d)(1)). Furthermore, as discussed in detail above, the claims as currently formulated are meaningful distinct from the improvement that that Desjardins instructs should be credited at §101.
As a result, the argument is not persuasive.
V. Applicant argues the rejections of the claims, under 35 USC § 103, should be withdrawn (Applicant’s Remarks, 03/27/2026, Pg. 15-18, Section “Section 35 U.S.C. § 103 Rejections”).
In response to Applicant’s amendments, the previously communicated rejections under 35 U.S.C. § 103, have been withdrawn. However, Applicants arguments are not persuasive in light of the new grounds for rejection, under 35 U.S.C. § 103, discussed in detail above. The new grounds of rejection rely on new prior art of record to teach the new combination of elements in the amended independent claims, which were not presented in this arrangement in any of the previously presented claims. As a result, Applicant’s arguments are rendered moot.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MATTHEW BRYCE GOLAN whose telephone number is (571)272-5159. The examiner can normally be reached Monday through Friday, 8:00 AM to 5:00 PM ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Alexey Shmatov can be reached at (571) 270-3428. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MATTHEW BRYCE GOLAN/Examiner, Art Unit 2123
/ALEXEY SHMATOV/Supervisory Patent Examiner, Art Unit 2123
Prosecution Insights