DETAILED ACTION
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 .
Response to Amendment
This office action is in response to the amendment filed on 06/26/2025. Claims remain pending in the application. Claims 1 and 15 are independent.
Response to Arguments
Applicant's arguments filed on 06/26/2026 have been fully considered but arguments with respect to FIG. 5 (see Page 3 of the Amendment to the Specification) and 101 rejections (see Pages 13-17 of the Remark) are not persuasive. Some of arguments with respect to the 103 rejection (see Pages 17-20 of the Remarks) are persuasive. The 103 rejection of Claims 1-7 and 12-14 are withdrawn.
Applicant argues on Page 3 of the Amendment to the Specification that Fig. 5 illustrates an example embodiment of the invention, while the cited sources referenced by the examiner appear to describe other example embodiments related to different Figures.
In response, examiner respectfully disagrees. The cited sources referenced in the previous Office Action are "If the element has a negative feedback, this means the element is probably a noise in the grouping, then the system can reduce the attribute weights by a decay factor conversely if the element has a positive feedback, this means the element is relevant to the grouping, then process 500 can increase the attribute weights by an increment factor. If the element was replaced (e.g. a remove followed by an addition to the same grouping) this means the element is noise and the added item was a substitute for it, then process 500 can increment the attribute weights of the differing attributes. Then process 500 can store the relevancy data back into the elements." which is quoted from ¶ [0075] of the specification. Therefore, ¶ [0075] and process 500 of FIG. 5 are the same embodiment, and step 1.6.6 "Increase relevancy attribute weights of all attributes by a decay" and step 1.6.8 "Increase relevancy attribute weights of all differing attributes by a decay" in process 500 of FIG. 5 are conflict with the description of ¶ [0075] as well as common knowledge in the art because "decay" has opposite meanings of "increase" (see 112 rejection). NOTE: step 1.6.5. "Decrease relevancy attribute weights of all attributes by a decay" in process 500 of FIG. 5 is consistent with the description of ¶ [0075] as well as common knowledge in the art because "decay" has similar meaning of "decrease".
Applicant further argues on Pages 13-17 of the Remarks regarding 101 rejection that THE CLAIMS ARE NOT DIRECTED TO AN ABSTRACT IDEA.
In response, examiner respectfully disagrees. To overcome 101 abstract idea rejection, additional elements/limitations (i.e., non-abstract idea elements/limitations) must be integrated with other abstract idea elements/limitations as a whole in a meaningful way (i.e., not just "apply it") so that the improvement of a particle technology/application described in the specification is reflected in the claim. The limitations "building a set of knowledge representations", "associating each individual elements of the plurality of elements into a grouping of elements using a plurality of attributes comprising an implicit attribute and an explicit attributes", "utilizing a plurality of knowledge sources and a plurality of complimentary knowledge sources that collected prior in an incremental manner" recited in the claims and mentioned in the Remarks all are mental processes. The "representing the plurality of relevancies into N-dimensional planes," "iteratively applying the filtering operation by increasing a subset of values and collecting a plurality of aggregations," "obtaining a statistical probability value of relevancy by comparing aggregations," "decreasing a relevancy attribute weights of the plurality of attributes by a decay on a negative feedback on elements," "increasing the relevancy attribute weights of the plurality of attributes by a decay on positive feedback on the plurality of elements," "finding a set of differing attributes and increasing the relevancy attribute weights of the set of differing attributes by a decay on substitution feedback on the plurality of elements," "extracting properties from the structure obtained and iterating amongst different planes of relevancies of source and destination elements," and "performing a selection technique to select the top relevant elements in the grouping by applying methods of ranking, threshold inferred from historical aggregations or user imputed domain expertise" recited in the claims and mentioned in the Remarks are either mathematical concepts/algorithms/calculations or mental processes. None of these limitations are additional elements/limitations (i.e., non-abstract idea elements/limitations), which cannot be used to integrated with other abstract idea limitations to form a practical application. .
Drawings
Applicant's amendment to the specification corrects only one of previous objections; therefore, only one of the previous objections are withdrawn. The remaining objections are shown below.
The drawings are objected to because (1) in 1.6.6 of FIG. 5, "Increase relevancy attribute weights of all attributes by a decay" is inconsistent with the specification in ¶ [0075] that "… if the element has a positive feedback, this means the element is relevant to the grouping, then process 500 can increase the attribute weights by an increment factor …"; (2) in 1.6.8 of FIG. 5, "Increase relevancy attribute weights of all differing attributes by a decay" is inconsistent with the specification in ¶ [0075] that "… If the element was replaced (e.g. a remove followed by an addition to the same grouping) this means the element is noise and the added item was a substitute for it, then process 500 can increment the attribute weights of the differing attributes …"; (3) in 704 of FIG. 7, "EXTRACT MATCHING ATTRIBUTES FROM THE AND ELEMENTS" appears to be "EXTRACT MATCHING ATTRIBUTES FROM THE USER AND ELEMENTS" (i.e., missing word between 'THE' and 'AND') according to ¶ [0077]; and (4) steps 802-808 in FIG. 8 (e.g., step 802: DERIVE THE IMPLICIT AND/OR EXPLICIT ATTRIBUTE FROM EACH OF ELEMENTS WHICH WAS PREVIOUSLY COLLECTED; step 804: ASSOCIATE THE INDIVIDUAL ELEMENTS INTO A GROUP OF (i.e., typo as 'FO' in FIG. 8) ELEMENTS USING THE IMPLICIT AND/OR EXPLICIT ATTRIBUTES; step 806: BOOTSTRAP KNOWLEDGE SOURCE EITHER INCREMENTALLY AND/OR USING PRIOR KNOWLEDGE SOURCE/COLLECT COMPLIMENTARY KNOWLEDGE BASES; step 808: ASSOCIATE THESE ELEMENTS TO THE EXISTING KNOWLEDGE SOURCES USING RELEVANCY TECHNIQUES) are completely different to the specification described in ¶ [0080] (e.g., collect all stored markings of relevant and/or irrelevant in step 802 for each element pair; aggregate the relevancy scoring positive value in step 804 for relevant pair and/or negative value in step 806 for irrelevant pair; perform a selection technique to select the top relevant elements in the grouping in step 808). Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Specification
Applicant's amendment to the abstract corrects only one of previous objections; therefore, only one of the previous objections are withdrawn. The remaining objections are shown below.
The disclosure is objected to because of the following informalities:
In ¶ [0003], "… a method of forming a plurality of elements with tagentiblity and hierarchy properties … wherein the plurality of elements are extracted from a domain information and are a meaning the domain information …" appears to be "… a method of forming a plurality of elements with tangibility and hierarchy properties … wherein the plurality of elements are extracted from a domain information and are a meaning of the domain information …" according to Claim 1;
the description in ¶ [0003] (similar to FIG.5) is conflicted with the description of ¶ [0075], and the common knowledge in the art because "increasing" and "decay" having opposite meaning (see also Drawing objection and 112 rejection);
In ¶ [0077], "Process 700 can extract the weights from the matching attributes which was optimized (e.g. see Figure 4, etc.)." appears to be "Process 700 can extract the weights from the matching attributes which was optimized (e.g. see Figure 5, etc.)." because according to ¶¶ [0007]-[0008], an example optimization process is described in Figure 5 and not described in Figure 4.
Appropriate correction is required.
Claim Objections
Applicant's amendment to claims corrects some of previous objections; therefore, some of previous objections are withdrawn. Applicant's amendment to claims also raises new issues; therefore, the remaining objections are shown below.
Claims 1 and 3-5 are objected to because of the following informalities:
in Claim 1, line 2, "… the method comprising the steps of: …" appears to be "… the method comprising steps of: …" (i.e., insufficient antecedent basis);
in Claim 1, line 13, "... associating the plurality of elements of claim into …" appears to be "... associating the plurality of elements into …" (moving from in previous 112(b) rejection);
in Claim 1, lines 17-18, "… decreasing a relevancy attribute weights of the plurality of attributes by a decay on a negative feedback on the plurality of elements …" appears to be "… decreasing relevancy attribute weights of the plurality of attributes by a decay for a negative feedback on the plurality of elements …";
in Claim 1, lines 19-20, "… increasing the relevancy attribute weights of the plurality of attributes by … on positive feedback on the plurality of elements …" appears to be "… increasing the relevancy attribute weights of the plurality of attributes by … for positive feedback on the plurality of elements …" (see also 112 rejections);
in Claim 1, lines 22-24, "… finding a set of differing attributes and increasing the relevancy attribute weights of the set of differing attributes by … on substitution feedback on the plurality of elements" appears to be "… finding a set of differing attributes and increasing the relevancy attribute weights of the set of differing attributes by … for substitution feedback on the plurality of elements";
in Claims 3-5, line 2, please clarify that "the set of knowledges sources", in the limitation "… iterating the plurality of elements over the set of knowledges sources by …", is referred to "the plurality of knowledges sources" or "the set of knowledges representations";
in Claim 4, lines 7-8, "… computing and tracking … on each pair of elements" appears to be "… computing and tracking … on each pair of the plurality of elements".
Appropriate correction is required.
Claim Rejections - 35 USC § 112
Applicant's amendment to claims corrects some of previous objections; therefore, some of previous objections are withdrawn. Applicant's amendment to claims also raises new issues; therefore, the remaining objections are shown below. Also, since Applicant described in Page 3 of the Amendment to the Specification that " Fig. 5 illustrates an example embodiment of the invention, while the cited sources referenced by the examiner appear to describe other example embodiments related to different Figures " without pointing out which paragraphs in the specification described details embodiment of FIG. 5, therefore, 112(a) the enablement requirement is raised instead of 112(a) the written description requirement assuming they are different embodiments (but actually ¶ [0075] and FIG. 5 cited in the previous Office Action are the same embodiment because ¶ [0075] of the specification explicitly describes details of process 500 in FIG. 5).
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-7 and 12-14 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the enablement requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to enable one skilled in the art to which it pertains, or with which it is most nearly connected, to make and/or use the invention.
The "… increasing the relevancy attribute weights of the plurality of attributes by a decay …" recited in Claim 1, lines 17-18 and "… increasing the relevancy attribute weights of the set of differing attributes by a decay …" recited in Claim 1, lines 22-23 are not disclosed in the specification in a manner clear enough and sufficient to enable a person skilled in the art to implement them. Claims 2-7 and 12-14 fully incorporates the deficiency of its base claims. The specification related to these limitations are the followings: (1) the specification in ¶ [0003] describes "… increasing the relevancy attribute weights of the plurality of attributes by a decay on positive feedback on the plurality of elements …" and "… increasing the relevancy attribute weights of the set of differing attributes by a decay on substitution feedback on the plurality of elements …" using the same claim language without further explanation how to achieve "increasing weights by a decay" when the "decay" has completely opposite meaning of the "increasing"; (2) step 1.6.6 of the process 500 shown in FIG. 5 describes "Increase relevancy attribute weights of all attributes by a decay" and step 1.6.8 of the process 500 in FIG. 5 "Increase relevancy attribute weights of all differ attributes by a decay" using similar claim language without any explanation in the specification how to achieve "increase weights by a decay" when the "decay" has completely opposite meaning of the "increase"; and (3) ¶ [0075] of the specification describes that "If the element has a negative feedback, this means the element is probably a noise in the grouping, then the system can reduce the attribute weights by a decay factor conversely if the element has a positive feedback, this means the element is relevant to the grouping, then process 500 can increase the attribute weights by an increment factor. If the element was replaced (e.g. a remove followed by an addition to the same grouping) this means the element is noise and the added item was a substitute for it, then process 500 can increment the attribute weights of the differing attributes. Then process 500 can store the relevancy data back into the elements." Hence, the ¶ [0003] of the speciation and steps 1.6.6 and 1.6.8 of the process 500 shown in FIG. 5 are conflicted with the details description of the process 500 in ¶ [0075] of the specification. Also, "… increase/increasing the attribute weights by an increment factor …" is recognized as common general knowledge in the art, and "… increase/increasing the attribute weights by a decay …" is not recognized as common general knowledge in the art because the "increment" has the similar meaning of the "increase/increasing" and the "decay" has the completely opposite meaning of the "increase/increasing". Therefore, it cannot be said that the specification discloses a method for "… increasing the relevancy attribute weights of the plurality of attributes by a decay …" and "… increasing the relevancy attribute weights of the set of differing attributes by a decay …". Furthermore, without details explanations of how to "increase" attribute weights by "a decay" in the specification, a person skilled in the art would not be able to repeat steps 1.6.6 and 1.6.8 of the process 500 shown in FIG. 5. Therefore, Claims 1-7 and 12-14 are not state clearly and sufficiently to enable a person skilled in the art to carry out the invention, and does not satisfy the enablement requirements.
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.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-7 and 12-17 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation " … associating each individual element of the plurality of elements into a grouping of elements …" in lines , the limitation " … removing a noise from a grouping of the plurality of elements … " in line 21, the limitation"… collecting the relevancy values of the grouping of the plurality of elements …" in line, and the limitation "… select a top relevant element in the grouping by …" in lines 34-35 , which rendering the claim indefinite because (1) it is unclear whether a/the "grouping of the plurality of elements" is the same or different to a "grouping of elements "since " each individual element of the plurality of elements" is associated with a "grouping of elements"; and (2) if they are different, which "grouping" is referred by "the grouping" when select a top relevant element. Clarification is required.
Claim 1 recites the limitation "… increasing the relevancy attribute weights of the plurality of attributes by a decay on positive feedback on the plurality of elements … finding a set of differing attributes and increasing the relevancy attribute weights of the set of differing attributes by a decay on substitution feedback on the plurality of elements …" in lines 1, which rendering the claim indefinite because it is unclear how can .
Claim 1 recites the limitation "… collecting the relevancy values of the grouping of the plurality of elements …" in line . There is insufficient antecedent basis for the limitation "the relevancy values" in the claim. Clarification is required.
Claim 1 recites the limitation "… representing the plurality of relevancies into N-dimensional planes …" in line . There is insufficient antecedent basis for the limitation "the plurality of relevancies" in the claim. Clarification is required.
Claim 1 recites the limitation "… extracting properties from the structure obtained …" in line . There is insufficient antecedent basis for the limitation "the structure" in the claim. Clarification is required.
Claim 1 recites the limitations " … representing the plurality of relevancies into N-dimensional planes … iterating amongst different planes of relevancies of source and destination elements … obtaining a statistical probability value of relevancy by comparing aggregations …" in lines , which rendering the claim indefinite because it is unclear (1) whether these instances of "relevancies" are the same or different (moving from previous claim objection to Claim 9); (2) that a "statistical probability value" of which "relevancy" among the "plurality of relevancies" is obtained. Clarification is required (see also ¶ [0093] of the specification).
Claim 1 recites the limitation "… iteratively applying a filtering operation by increasing a subset of values and collecting a plurality of aggregations, obtaining a statistical probability value of relevancy by comparing aggregations … " in lines 29-32, which rendering the claim indefinite because it is unclear whether these instances of "aggregations" are the same or different. Clarification is required.
Claims 2-7 and 12-14 are rejected for fully incorporating the deficiency of their respective base claims.
Claim 2 recites the limitation "the relevancy data" in line 2. There is insufficient antecedent basis for this limitation in the claim. Clarification is required.
Claims 3-7 are rejected for fully incorporating the deficiency of their respective base claims.
Claim 4 recites the limitation "… computing and tracking the relevancy and complimenting relevancy score on each pair of elements" in line. There is insufficient antecedent basis for the limitation "the relevancy and complimenting relevancy score" in the claim. Clarification is required (see also ¶ [0067] of the specification).
Claims 5-7 are rejected for fully incorporating the deficiency of their respective base claims.
Claim 5 recites the limitation "… reversing the association direction once the destination element" in line , which rendering the claim indefinite because (1) there is insufficient antecedent basis for the limitations "the association direction" and "the destination element" in the claim; and (2) it seems that th.
Claims 6-7 are rejected for fully incorporating the deficiency of their respective base claims.
Claims 6 recites the limitation "... determining the relevancy between the plurality of elements by: ..." in line 2, which rendering the claim indefinite because ".
Claim 6 recites the limitation "the matching attributes" in line 3. There is insufficient antecedent basis for this limitation in the claim. Clarification is required.
Claim 6 recites the limitation "... extracting the matching attributes and relevancy weights defined from the plurality of elements … by utilizing the attribute weights " in line, which rendering the claim indefinite because ".
Claim 6 recites the limitation "the user of the system" in lines 4-5. There is insufficient antecedent basis for the limitations "the user" and "the system" in the claim. Clarification is required.
Claim 6 recites the limitation "… applying the relevancy technique defined for the particular type by …" in line 7. There is insufficient antecedent basis for the limitation "the particular type". Clarification is required.
Claim 7 is rejected for fully incorporating the deficiency of their respective base claims.
Claim 7 recites the limitation "... determining the relevancy between the plurality of elements by: storing the relevancy values … determine the relevancy by extracting the relevancy value … storing the relevancy value" in lines , which rendering the claim indefinite because "… collecting the relevancy values of the grouping of the plurality of elements … representing the plurality of relevancies … iterating amongst different planes of relevancies of source and destination elements … obtaining a statistical probability value of relevancy by comparing aggregations…" is recited in its based Claim 1, "… storing the relevancy data back into the knowledge sources …" is recited in its based Claim 2, "… computing and tracking the relevancy and complimenting relevancy score on each pair of elements" is recited in its based Claim 4, and "... determining the relevancy between the plurality of elements by: ..." is recited in its based Claim 6, and it is unclear whether "the relevancy between the plurality of elements", "the relevancy values", "the relevancy", and "the relevancy value" recited here are referred to which instance of "relevancy", "relevancies", or "relevancy values" cited in its based claim. Clarification is required.
Claim 13 (depending on Claims 1 and 12) recites the limitation "wherein the different knowledge backings are the representations of the same element in different domains" in line 1. There is insufficient antecedent basis for the limitations "the different knowledge backings" and "the representations of the same element" in the claim. Clarification is required. NOTE: Although "a set of different knowledge backings" and "one or more representations of a same element in different domains" are recited in Claim 6, the limitations of Claim 13 is completely included in Claim 6 and therefore, it cannot be changed to depend on Claim 6 for resolving the antecedent basis issues.
Claim 14 is rejected for fully incorporating the deficiency of their respective base claims.
Claim 14 recites the limitation "wherein the different knowledge backings are used to determine the relevancy with respect to the user considering the various different types of knowledge representation in the head of different users" in lines 1-3. There is insufficient antecedent basis for the limitations "the relevancy with respect to the user", "the user", "the various different types of knowledge representation" and "the head of different users" in the claim. Clarification is required.
Claim 15 recites the limitation "A computer-implemented method for grouping elements, the method comprising: receiving a plurality of elements for a grouping operation …" in lines 1-3, the limitation "… determining, for each pair of elements …" in line 8, the limitation "… modify attribute weights of the elements …" in lines 14-15, and the limitation "… ranking the elements based on …" in line 18, which rendering the claim indefinite because it is unclear whether the first three instances of "elements" are the same or different and if they are different which instance of "elements" is refereed by the last two instances of "the elements". Clarification is required.
Claim 15 recites the limitation "… the relevancy score incorporating contextual factors and feedback … updating the relevancy score based on user feedback …" in lines 10-13, which rendering the claim indefinite because it is unclear whether these two instances of "feedback" are the same or different. Clarification is required.
Claim 15 recites the limitation "… determining, for each pair of elements from the pairwise combinations, a relevancy score based on … wherein relevance is aggregated across multiple association dimensions … updating the relevancy score based on user feedback … computing an aggregate relevancy score for each element in the plurality of elements based on the updated relevancy score…" in lines 8-17, which rendering the claim indefinite because (1) it is unclear whether "relevance" aggregated before updating is the same or different to "relevancy score" for each pair of elements before updating; and (2) usually a relevancy score is for a pair of elements, and it is unclear how can an aggregate relevancy score for each element is computed without considering a pair of elements. Clarification is required.
Claims 16-17 are rejected for fully incorporating the deficiency of their respective base claims.
Claim 16 recites the limitation "… determining the relevancy score for a pair of elements …" in lines 1-2, which rendering the claim indefinite because "… method for grouping elements, the method comprising: receiving a plurality of elements for a grouping operation … determining, for each pair of elements …" is also recited in its based claim and it is unclear whether these instances of "elements" are the same or different. Clarification is required.
Claim 17 recites the limitation "… wherein the relevancy score is computed by evaluating contextual association pathways between attributes of the paired elements … ", which rendering the claim indefinite because "… determining, for each pair of elements from the pairwise combinations, a relevancy score based on …" is also recited in its based claim and it is unclear "the paired elements" is referred to "elements in a pair", "elements in each pair" or "elements in all pair". Clarification is required.
NOTE: terms referred to the same item seems to be used inconsistently throughout the claims, please make sure that terms referred to the same item are used consistently.
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 1-7 and 12-14 are rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract idea without significantly more. The claim(s) recite(s) ".
This judicial exception is not integrated into a practical application for Claims 1-14 because th.
Claims 1-7 and 12-14 do not include additional elements that are sufficient to amount to significantly more than the judicial exception because t.
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 15 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Potts et al. (US 2022/0129766 A1, filed on 06/22/2021), hereinafter Potts in view of BHATTACHARYYA et al. (US 2021/0065091 A1, pub. date: 03/04/2021), hereinafter BHATTACHARYYA and MCINTRE et al. (US 2009/0248672 A1, pub. date on 10/01/2009), hereinafter MCINTRE.
Independent Claim 15
Potts discloses a computer-implemented method for grouping elements (Potts, ¶ [0002]: organize data into discrete sets and using standardized identifiers, categories, and characteristics; ¶ [0041]: namespaces are commonly structured as hierarchies to allow reuse of names in different contexts; ¶¶ [0042] and [0079]: an "entity type" refers to a particular category or label for a given entity or multiple entities sharing at least one common aspect; ¶¶ [0047] and [0070]: as with entities, nodes may be categorized according to different node types, and a given node may be associated with one or more attributes; ¶¶ [0048] and [0074]: as with nodes, edges may be categorized according to different types (i.e., of relationships), and a given edge may be associated with a unique primary identifier and one or more attributes; ¶ [0050]: an "ontology" refers to a definition, naming, and representation of categories and properties of entities, and relationships between entities, pertaining to a particular information domain, including subdomains and/or overlapping domains (this is sometimes referred to as a "domain ontology"); ¶¶ [0059]-[0060]: machine learning tasks conventionally are classified into multiple categories; unsupervised learning algorithms are used to find structure in the data, like grouping or clustering of data points; unsupervised learning can discover patterns in the data, and can group the inputs into categories, as in "feature learning"), the method comprising:
receiving a plurality of elements for a grouping operation (Potts, ¶ [0034]: one or more files in a dataset may include, for example, data that was generated by the source, data that was collected by the source, data that was received by the source, and/or data that was generated, collected and/or received by the source and modified or curated in some manner by the source; ¶ [0041]: a dataset received from a particular source is stored in a namespace associated with the particular source; ¶¶ [0057]-[0060]: artificial intelligence must have access to information regarding various entities objects, categories, properties) and relationships between entities, to implement knowledge engineering; Machine Learning-Machine learning (ML) is a branch of artificial intelligence based on the idea that systems (e.g., intelligent agents) can learn from data, identify patterns and make decisions with minimal human intervention; machine learning tasks conventionally are classified into multiple categories; unsupervised learning algorithms are used to find structure in the data, like grouping or clustering of data points; unsupervised learning can discover patterns in the data, and can group the inputs into categories, as in "feature learning");
creating all pairwise combinations from the plurality of elements (Potts, ¶¶ [0068]-[0069] and [0097] with FIG. 1: a given dataset generally includes information relating to one or more "entities" (things) having particular "entity types" ( categories or labels for entities sharing at least one common aspect) that pertain to the domain(s) of interest for which the RKG 100 is constructed and maintained; the first subgraph 150A of the example RKG 100 represents a first dataset including information relating to the entity type "diseases"; the second subgraph 150B represents a second dataset including information relating to the entity type "drugs"; each node in the RKG 100 represents an entity having a particular entity type, each edge represents a relationship of a particular type between two entities, and a graph schema for a given subgraph specifies types for nodes and edges (e.g., corresponding to types of entities and relationships), and a particular arrangement of nodes and edges based on the entities and relationships represented in the corresponding dataset; the RKG 100 illustrated in FIG. 1 may be created and maintained using a graph database management system; suitable file formats and database management systems for an RKG pursuant to the concepts disclosed herein allow for 1) various node types, 2) various edge types, 3) directed edges, 4) node and edge attributes having at least the types "string," "integer," "float," and lists thereof, and 5) multiple edges between pairs of nodes; ¶ [0091] with FIG. 2: for a set of n nodes representing (or deemed to represent) the same entities in different subgraphs, the number of edges needed to directly and completely connect respective pairs of the n nodes between the different subgraphs is given by the binomial coefficient; e.g., considering an example in which there are ten different subgraphs each containing the node "Disease 1," according to the binomial coefficient above 45 edges would be required to pairwise interconnect these nodes directly (10 choose 2=45); ¶ [0100]: as an initial matter, the information domain(s) for which an RKG is desired is/are first specified, such that multiple datasets from one or more sources may be preliminarily identified that are available and germane to the domain( s) of interest; ¶¶ [0122]-[0138] with FIGS. 3-4: available datasets pertaining to the domain(s) of interest may be respectively downloaded (e.g., from the Internet) and imported into corresponding isolated namespaces of computer storage (which namespaces may be labeled, based at least in part, on the source of the dataset); thereafter, a given dataset may be processed so as to generate a subgraph representing the dataset; a "graph schema" is created for the dataset to define the node types and the edge types that are used in the subgraph to represent the dataset; ¶¶ [0141]-[0143] with FIG.6: the method of FIG. 6 is performed on a subgraph-by-subgraph basis and may be performed sequentially on a number of subgraphs in succession or contemporaneously on multiple subgraphs; in block 610, a first node type is selected in the subgraph under consideration; if this first node type is not a canonical node type, as illustrated in blocks 620, 630 and 640 the method then proceeds to the next node type in the subgraph; if there are no more node types remaining for consideration, the method ends; if however the node type presently under consideration is a canonical node type, in block 650 consider if there are already nodes of this type in the canonical layer of the RKG; if not, in block 660 all of the nodes of this type and any edges coupled to these nodes are copied from the subgraph into the canonical layer, and in block 680 edges of the type "IS" are run between respective pairs of corresponding nodes in the canonical layer and the subgraph; if in block 650 it is determined that there are already canonical nodes of the type in question in the canonical layer, in block 670 consider if the number of canonical nodes of this type already present in the canonical layer is less than the number of subgraph nodes of this type; if not (i.e., if the set of canonical nodes of the type in question is a superset of the subgraph nodes of the same type), proceed to block 680 and runs edges of the type "IS" between respective pairs of corresponding nodes in the canonical layer and the subgraph; in block 670 of FIG. 6, if the number of canonical nodes of the type in question is less than the number of subgraph nodes of the same type (the set of subgraph nodes of the type in question is a superset of the canonical nodes of this type), then in block 690 those subgraph nodes of the type in question that are not already in the canonical layer ("delta nodes"), as well as any edges connected to these nodes, are copied into the canonical layer as canonical nodes and edges; in an alternative implementation of block 690, the entire set of subgraph nodes of the type in question (and their corresponding edges) may be copied into the canonical layer and thereby replace any preexisting canonical nodes of this type; then proceed to block 680 where edges of the type "IS" are run between respective pairs of corresponding nodes in the canonical layer and the subgraph; once edges of the type "IS" are run between the corresponding nodes of the type in question, proceed to block 630 to see if there are any remaining node types in the subgraph to consider for possible addition to the canonical layer; the method ends when all node types in the subgraph have been thusly considered); deriving, for each element of the plurality of elements, a set of attributes based on a domain-specific knowledge backing constructed from a general-purpose knowledge representation (Potts, ¶ [0045]: an "attribute" is an identifier, aspect, quality, or characteristic of an entity or a relationship; ¶¶ [0050]-[0051]: an ontology is typically based on logical formalisms that support some form of inference in connection with available data pertaining to the information domain(s), and thereby allows implicit information to be derived from the available explicit data; ontologies have been created for some information domains to reduce complexity and organize knowledge and data in the domain(s); this in turn improves communication about the domain(s), and analysis of data and problem solving in the domain(s); a knowledge graph encodes the meaning of the data that it represents (e.g., by using node and edge identifiers, types and attributes that are familiar to those interested in, or practitioners of, the information domain); a knowledge graph may be queried in a style that is closer to a natural language (e.g., by virtue of the ontologies employed, which would include vocabulary familiar to practitioners in the domain of interest); this facilitates search and discovery of information encoded in the knowledge graph; characteristics pertaining to both nodes and edges in a knowledge graph (e.g., identifiers, types, attributes associated with nodes and edges) may be subjected to computer analytical operations (e.g., being passed as an argument, returned from a function modified, and assigned to a variable); ¶¶ [0091]-[0097] with FIGS. 1-2: the canonical layer provides for a substantial reduction of graph complexity (e.g., number of edges) required to interconnect respective corresponding nodes in different subgraphs; within the canonical layer of an RKG, a given canonical node may be connected to one or more other canonical nodes via respective edges of a wide variety of types, based at least in part on the diverse relationships that may exist between canonical nodes of the same type or different types; e.g., as shown in FIG. 1, the canonical node 124A ("Drug 2") is connected via an edge of the type "TREATS" to the canonical node 122C ("Disease 3"); similarly, the canonical node 124B ("Drug 1") is connected via an edge of the type "TREATS" to the canonical node 122B ("Disease 2"); edges between subgraph nodes and canonical nodes, or between any two canonical nodes, may be generated based at least in part on: 1) one or more particular attributes of the respective nodes, 2) relationships between entities specified in some manner by the underlying information in the datasets represented by the subgraphs of the RKG, and/or 3) trained models that predict (based on a variety of criteria coded in logic for the model) that the nodes should be connected as having some particular type of articulated relationship (with some corresponding probability); connect subgraph node of type X to canonical node of type X with an edge of type "IS" if the respective primary identifiers of the nodes match; connect subgraph node of type Y to canonical node of type Y with an edge of type "IS" if respective attributes A1, A3 and A5 have the same values for the respective nodes; an edge may be generated between a subgraph node and a canonical node, or between two canonical nodes, based on a trained model ( also referred to herein further below as a "model-based connector") that predicts in some respect the relationship between the nodes; the degree of certainty may be recorded as a probability attribute of the edge of type "IS" (e.g., using a number from O to 1, inclusive); ¶¶ [0155]-[0159]: consider a new dataset for addition to the RKG; extract entity types, entities, and attributes for entities that are deemed to be relevant in some manner to the new dataset, and these may be organized in tabular form; relevant information extracted from the existing RKG and the new dataset are represented as two tables ( e.g., in which the column headers for the respective tables may represent in some manner one or more entity types included in the table, and in which respective rows in each table include values for the entities of the types represented by the column headers); the process of designing a model-based connector to connect nodes of a subgraph representing the new dataset to sufficiently corresponding nodes in the canonical layer may employ "active learning"; to this end, human annotators would be presented with pairs of entries from each of the two tables and asked to say "Yes, these rows respectively refer to the same person" or "No, these rows respectively refer to different people"; once the model is performing at an acceptable confidence level, it can then be deployed on the entire new dataset to predict corresponding nodes with sufficient certainty and generate edges of the type "IS" between such pairs of nodes (with the uncertainty recorded as an attribute of the edge of the type "IS"); ¶ [0188]: the language models used for entity detection are trained on 'name'-type attributes of nodes, and resolving those entities is graph-backed: the 'Entity index' is automatically created from the database and provides fast look-up);
determining, for each pair of elements from the pairwise combinations, a relevancy score based on a distance metric evaluated over a domain-specific structure, the relevancy score incorporating contextual factors and feedback (Potts, ¶¶ [0156]-[0159]: extract entity types, entities, and attributes for entities that are deemed to be relevant in some manner to the new dataset, and these may be organized in tabular form; relevant information extracted from the existing RKG and the new dataset are represented as two tables (e.g., in which the column headers for the respective tables may represent in some manner one or more entity types included in the table, and in which respective rows in each table include values for the entities of the types represented by the column headers); human annotators would be presented with pairs of entries from each of the two tables and asked to say "Yes, these rows respectively refer to the same person" or "No, these rows respectively refer to different people" (i.e., feedback); an ML model ( e.g., for a classifier) may be developed for the model-based connector and trained on the initial human annotations; a feature function be created ("featurization") which is run on raw inputs (in the current example, table rows) to obtain purely numerical representations ( e.g., degrees of certainty regarding a possible match); the existing RKG itself may be used to build such feature functions; e.g., the existing RKG might be used to obtain the 'Specialty distance' values, which indicate how far apart two specialties are in the canonical taxonomy of medical specialties); similarly, the existing RKG may be useful in getting a distance estimate between two zip codes, in normalizing place and entity names, and in doing more sophisticated name comparisons (e.g., the likelihood of the name Zoltan Kim given the likelihoods of Zoltan as a first name and Kim as a last name); learning to weight the features in the above table to maximize the likelihood of the human annotated examples; once the model is performing at an acceptable confidence level, it can then be deployed on the entire new dataset to predict corresponding nodes with sufficient certainty and generate edges of the type "IS" between such pairs of nodes (with the uncertainty recorded as an attribute of the edge of the type "IS"));
;
updating the relevancy score (Potts, ¶ [0051]: new data items or datasets may be added to a knowledge graph over time; in particular, one or more ontologies on which the knowledge graph is based may be extended and/or revised as new data is considered for addition to the graph, and new entities and/or entity types in datasets may be represented as nodes and connected via edges to existing nodes (based on existing or extended/revised ontologies); this makes know1edge graphs convenient for storing and managing data in use cases where regular data updates and/or data growth are important, particularly when data is arriving from diverse, heterogeneous sources; ¶¶ [0063] and [0100]: a Roam Knowledge Graph (RKG) is an effective and highly useful structure for storing and managing data for a variety of use cases and provides specific advantages particularly when data is dynamic (e.g., where regular data updates and/or data growth are important) and when data is heterogeneous and arriving from diverse sources; as a general premise, an RKG has notable utility in providing links between two or more datasets, particularly when one or more of the datasets includes dynamic information (e.g., where regular data updates and/or data growth are important) and when the datasets are heterogeneous and arriving from diverse sources; ¶ [0120]: as one or more of the initial datasets on which an RKG is based are updated and/or evolve over time, and/or as one or more new datasets are identified (and stored in one or more new isolated namespaces) to be represented as subgraphs and connected to the canonical layer of an RKG, new entity types in the datasets may be identified as appropriate candidates for augmenting the canonical layer with additional canonical node types (e.g., based on various criteria similar to those discussed above));
computing an aggregate relevancy score for each element in the plurality of elements based on the (Potts, ¶¶ [0036], [0067], and [0101]: "structured data" refers to multiple data elements than can be meaningfully aggregated, and that generally are organized into a formatted repository of data elements (e.g., a spreadsheet or database including one or more tables with rows and columns), so that respective elements of the data are addressable and easily accessible and searchable (e.g., for processing and