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 .
Remarks
This action is in response to the application received on 5/22/26. Claims 1-20 are pending in the application.
Claims 1-20 are rejected under 35 U.S.C. 101.
Claims 1 and 11-20 are rejected under 35 U.S.C. 103 as being unpatentable over Yang et al. (US 12,579,179), and further in view of Simard et al. (US 2023/0315773).
Claims 2 and 5-10 are rejected under 35 U.S.C. 103 as being unpatentable over Yang in view of Simard, and further in view of Peraud et al. (US 2022/0230089).
Claims 3 and 4 are rejected under 35 U.S.C. 103 as being unpatentable over Yang in view of Simard and Peraud, and further in view of Lewis et al. (US 9,367,814).
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 an abstract idea without significantly more.
Step 2A, Prong One asks: Is the claim directed to a law of nature, a natural phenomenon (product of nature) or an abstract idea? See MPEP 2106.04 Part I. 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. See MPEP 2106.04(a).
With respect to claims 1, 19, and 20, the limitation of “generating, via the one or more processors, a request for documents from the corpus of documents based on the first issue or the new issue” and “generating, via the one or more processors, a prompt based on the prompt criteria”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting “via the one or more processors,” nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the “via the one or more processors” language, “generating” in the context of this claim encompasses the user mentally determining data that is needed. Similarly, the limitation of “classifying, via the one or more processors, documents within the initial set of documents by inputting the prompt and the documents within the initial set of documents into the generative Al model” and “classifying, via the one or more processors, the obtained documents by inputting the prompt and the obtained documents into the generative Al model”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the “via the one or more processors” language, “classifying” in the context of this claim encompasses the user mentally determining a group of similar documents. The limitation of “evaluating, via the one or more processors, classification performance of the prompt” and “evaluating, via the one or more processors, the classified obtained documents against ground truth data associated with the obtained documents to determine that the first issue or the new issue has been addressed”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the “via the one or more processors” language, “evaluating” in the context of this claim encompasses the user mentally determining if the initial data for forming groups of documents is sufficient. 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. Accordingly, the claim recites an abstract idea.
At step 2a, prong two, this judicial exception is not integrated into a practical application. The recite a processor to execute the operations and a generative artificial intelligence model, however, these are recited as a high-level of generality (i.e., as a generic processor performing a generic computer function) such that it amounts to no more than mere instructions to apply the exception using generic computer components. Additionally, the claim recites “obtaining, via one or more processors, an initial set of documents, “obtaining, via the one or more processors, prompt criteria,” “obtaining, via the one or more processors, documents responsive to the request for documents,” and “adding, via the one or more processors, the obtained documents to the initial set of documents.” These elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception and provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception using generic computer components. Mere instructions to apply an exception using generic computer components cannot provide an inventive concept.
With respect to “obtaining, via one or more processors, an initial set of documents, “obtaining, via the one or more processors, prompt criteria,” and “obtaining, via the one or more processors, documents responsive to the request for documents,”, the courts have found limitations directed towards data gathering to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information).
With respect to “adding, via the one or more processors, the obtained documents to the initial set of documents”, the courts have found limitations directed towards storing to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Electronic recordkeeping, Alice Corp. Pty. Ltd. v. CLS Bank Int'l, 573 U.S. 208, 225, 110 USPQ2d 1984 (2014) (creating and maintaining "shadow accounts") and “storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015).
Considering the additional elements individually and in combination and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. The claim is not patent eligible.
With respect to claims 2, 4, 10, and 16, the limitations further define elements of the mental process and do not impose a meaningful limit on the judicial exception.
With respect to claim 3, the limitations directed towards “identifying” and “generating” are further mental process steps that encompass the user analyzing data and determining documents to gather. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
With respect to claims 5, 7, 8, 11, 12, 15, and 17, the limitations are directed towards further mental steps by “evaluating” data which encompasses the user analyzing data. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
With respect to claims 6, 9, and 18, the limitations directed towards “identifying” are further mental process steps that encompass the user analyzing data and determining documents to gather. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
With respect to claim 13, the limitations directed towards “obtaining” and “applying” provide additional elements, however, these elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception. They provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea.
With respect to “obtaining,” the courts have found limitations directed towards data gathering to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information).
With respect to “applying”, the courts have found limitations directed towards storing to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Electronic recordkeeping, Alice Corp. Pty. Ltd. v. CLS Bank Int'l, 573 U.S. 208, 225, 110 USPQ2d 1984 (2014) (creating and maintaining "shadow accounts") and “storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015).
With respect to claim 14, the limitations directed towards “classifying” are further mental process steps that encompass the user mentally determining a group of similar documents. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Claim Rejections - 35 USC § 103
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 and 11-20 are rejected under 35 U.S.C. 103 as being unpatentable over Yang et al. (US 12,579,179), and further in view of Simard et al. (US 2023/0315773).
With respect to claim 1, Yang teaches a computer-implemented method for using a generative artificial intelligence (Al) model to classify documents, the computer-implemented method comprising:
obtaining, via one or more processors, an initial set of documents from a corpus of documents (Yang, Col. 19 Li. 56-58, the process 400 begins at step/operation 402 when the document analysis computing entity 106 receives one or more document data objects.);
obtaining, via the one or more processors, prompt criteria defining an inquiry associated with the corpus of documents, wherein the prompt criteria define one or more issues associated with the corpus of documents (Yang, Col. 22 Li. 17-21, A prompt command may comprise a context to analyze comprising one or more examples, one or more analysis inputs, and a task-specific instruction to follow and repeat based on the one or more examples and the one or more analysis inputs);
generating, via the one or more processors, a prompt based on the prompt criteria (Yang, Col. 22 Li. 4-8, assigning a model-assigned categorical identifier comprises either generating or receiving one or more prompt commands);
classifying, via the one or more processors, documents within the initial set of documents by inputting the prompt and the documents within the initial set of documents into the generative Al model (Yang, Col. 22 Li. 4-8, assigning a model-assigned categorical identifier comprises … providing the one or more prompt commands to a generative machine learning model. & Col. 22 Li. 28-34, As depicted in FIG. 6, prompt command 600 comprises an example 602, analysis inputs 604, and task-specific instructions 606. Example 602 comprises a document data object or one or more text segment data objects, and categorical identifiers assigned to the document data object or one or more text segment data objects.);
evaluating, via the one or more processors, classification performance of the prompt to identify (i) that the initial set of documents does not include enough documents associated with a first issue of the one or more issues, or (ii) that the corpus of documents is associated with a new issue (Yang, Col. 23 Li. 33-37, at step/operation 508, the document analysis computing entity 106 generates, using a verifier machine learning model, one or more evidence predictions on respective ones of the one or more evidence text portions.).
Yang doesn't expressly discuss generating, via the one or more processors, a request for documents from the corpus of documents based on the first issue or the new issue; obtaining, via the one or more processors, documents responsive to the request for documents; adding, via the one or more processors, the obtained documents to the initial set of documents, classifying, via the one or more processors, the obtained documents by inputting the prompt and the obtained documents into the generative Al model; and evaluating, via the one or more processors, the classified obtained documents against ground truth data associated with the obtained documents to determine that the first issue or the new issue has been addressed.
generating, via the one or more processors, a request for documents from the corpus of documents based on the first issue or the new issue (Simard, pa 0062-0063, the Add Document Assistant may include the following functions: [0063] (i) Select Classification(s), which displays a list of available classification trees in the content server. This enables the user to select individual classifications or to select an entire classification tree, and to import all of the documents with that classification assigned.);
obtaining, via the one or more processors, documents responsive to the request for documents (Simard, pa 0062-0063, the Add Document Assistant may include the following functions: [0063] (i) Select Classification(s), which displays a list of available classification trees in the content server. This enables the user to select individual classifications or to select an entire classification tree, and to import all of the documents with that classification assigned. The auto-classification system can be configured to retrieve documents with the specified classification from every location in the content server.);
adding, via the one or more processors, the obtained documents to the initial set of documents (Simard, pa 0062, the system permits the user to add more exemplars. This may, for example, be done in response to the guide presenting a recommended action to the user to add more exemplars.);
obtaining, via the one or more processors, documents responsive to the request for documents (Simard, pa 0063, The auto-classification system can be configured to retrieve documents with the specified classification from every location in the content server. Content may also be imported into the auto-classification system from any other content source or repository, e.g., content management systems (CMS), file systems, etc.);
classifying, via the one or more processors, the obtained documents by inputting the prompt and the obtained documents into the generative Al model (Simard, pa 0064, Sampling Settings, which, when enabled, allow the system to select a random set of documents from the set of documents being imported, and convert them to test documents. Test documents are used to test the model's accuracy before creating an actual classification job & pa 0079, A test run may be performed on a small training corpus of documents that the user selects for this purpose. Based on a review collection, the system displays classification metrics on a metric panel from this test run to provide the user with feedback on the accuracy of the model.); and
evaluating, via the one or more processors, the classified obtained documents against ground truth data associated with the obtained documents to determine that the first issue or the new issue has been addressed (Simard, pa 0080, The review collection is a set of non-classified documents from a content server on which a classification test is run. A review collection enables the user to compare the accuracy of a classification assigned by the auto-classification system (by applying the model) with the classification that the user would assign manually).
It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified Yang with the teachings of Simard because it enhances the classification accuracy of the model (Simard, pa 0040).
With respect to claim 11, Yang in view of Simard teaches the computer-implemented method of claim 1, wherein evaluating classification performance of the prompt includes: generating, via the one or more processors, one or more respective statistical metrics for each of the one or more issues, wherein the one or more respective statistical metrics each include one or more of: an accuracy metric, a precision metric, a recall metrics, or an elusion metric (Simard, pa 0045, The auto-classification system (auto-classifier) is configured to provide feedback to the user in the form of metrics or other information that the user can review to determine whether the model is sufficiently accurate and for optionally adjusting the classification model to improve its accuracy. & pa 0047-0052, specific metrics).
With respect to claim 12, Yang in view of Simard teaches the computer-implemented method of claim 11, wherein identifying that the initial set of documents does not include enough documents associated with the first issue of the one or more issues (Simard, pa 0086-0087, This guide 630 presents recommendations 632, 634 or recommended actions, suggestions, tips, or other information that guides the user in the process of refining the classification model to obtain better results. As shown in this example, a first recommendation 632 is presented along with a second recommendation 636. [0087] As illustrated by way of example in FIG. 6, recommendation #1 suggests “YOU DO NOT HAVE ENOUGH DOCUMENTS FOR A STATISTICALLY VALID SAMPLE TO OBTAIN YOUR DESIRED CONFIDENCE LEVEL.” A user interface element 634 is displayed to enable the user to remedy the deficiency by clicking on the recommended action (e.g., “CLICK HERE TO ADD MORE DOCUMENTS”).) comprises: evaluating, via the one or more processors, one or more statistical metrics of the one or more respective statistical metrics associated with the first issue to identify that at least one statistical metric is statistically insignificant based on the amount of documents of the initial set of documents associated with the first issue (Simard, pa 0081, the dynamic user feedback guide may be activated and displayed only when the accuracy falls below a predetermined threshold. For example, if the recall and/or precision values are below a predetermined threshold, the guide may be activated and displayed onscreen. ).
With respect to claim 13, Yang in view of Simard teaches the computer-implemented method of claim 1, wherein evaluating classification performance of the prompt comprises: obtaining, via the one or more processors, review data associated with the initial set of documents including ground truth data associated the one or more issues; and applying, via the one or more processors, the review data to determine classification performance of the prompt with respect to the one or more issues (Simard, pa 0080, The review collection is a set of non-classified documents from a content server on which a classification test is run. A review collection enables the user to compare the accuracy of a classification assigned by the auto-classification system (by applying the model) with the classification that the user would assign manually).
With respect to claim 14, Yang in view of Simard teaches the computer-implemented method of claim 1, further comprising: classifying, via the one or more processors, the obtained documents responsive to the request for documents by inputting the prompt and the obtained documents into the generative Al model (Simard, pa 0064, Sampling Settings, which, when enabled, allow the system to select a random set of documents from the set of documents being imported, and convert them to test documents. Test documents are used to test the model's accuracy before creating an actual classification job & pa 0079, A test run may be performed on a small training corpus of documents that the user selects for this purpose. Based on a review collection, the system displays classification metrics on a metric panel from this test run to provide the user with feedback on the accuracy of the model.).
With respect to claim 15, Yang in view of Simard teaches the computer-implemented method of claim 14, further comprising: evaluating, via the one or more processors, classification performance of the prompt based on ground truth data associated with the obtained documents (Simard, pa 0080, The review collection is a set of non-classified documents from a content server on which a classification test is run. A review collection enables the user to compare the accuracy of a classification assigned by the auto-classification system (by applying the model) with the classification that the user would assign manually).
With respect to claim 16, Yang in view of Simard teaches the computer-implemented method of claim 1, wherein the one or more issues are associated a knowledge graph of facts (Yang, Col. 14 Li. 4-7, "categorical description" refers to a data construct that describes one or more keywords, numbers, or phrases associated with a classification according to a given taxonomy. Examiner note: a taxonomy serves the same purpose here as a knowledge graph and are both well known elements for storing data).
With respect to claim 17, Yang in view of Simard teaches the computer-implemented method of claim 16, further comprising: evaluating, via the one or more processors, the knowledge graph of facts and the prompt criteria to identify the new issue (Yang, Col. 14 Li. 10-14, one or more categorical identifiers associated with the one or more categorical descriptions are assigned to a document data object or one or more text segment data objects associated with the document data object. & Col. 22 Li. 4-8, assigning a model-assigned categorical identifier comprises … providing the one or more prompt commands to a generative machine learning model. & Col. 22 Li. 28-34, As depicted in FIG. 6, prompt command 600 comprises an example 602, analysis inputs 604, and task-specific instructions 606. Example 602 comprises a document data object or one or more text segment data objects, and categorical identifiers assigned to the document data object or one or more text segment data objects.).
With respect to claim 18, Yang in view of Simard teaches the computer-implemented method of claim 16, wherein identifying that the initial set of documents does not include enough documents associated with a first issue of the one or more issues (Simard, pa 0086-0087, This guide 630 presents recommendations 632, 634 or recommended actions, suggestions, tips, or other information that guides the user in the process of refining the classification model to obtain better results. As shown in this example, a first recommendation 632 is presented along with a second recommendation 636. [0087] As illustrated by way of example in FIG. 6, recommendation #1 suggests “YOU DO NOT HAVE ENOUGH DOCUMENTS FOR A STATISTICALLY VALID SAMPLE TO OBTAIN YOUR DESIRED CONFIDENCE LEVEL.” A user interface element 634 is displayed to enable the user to remedy the deficiency by clicking on the recommended action (e.g., “CLICK HERE TO ADD MORE DOCUMENTS”).) comprises: identifying, via the one or more processors, one or more regions of the knowledge graph of facts associated with one or more misclassifications of documents of the initial set of documents and the first issue of the one or more issues (Peraud, pa 0142, a human may define a taxonomy of known topics that she wishes to infer, track, and monitor. The set of classes may include sets of seed datasets belonging to the set of classes. The seed datasets may have ground-truth labels (or definite labels) corresponding to the set of classes, assigned by the SME. & pa 0145, In act 610, pseudo labels corresponding to the classes may be assigned to the raw datasets based on the distances calculated in act 608.).
With respect to claims 19 and 20, the limitations are essentially the same as claim 1, and are rejected for the same reasons.
Claims 2 and 5-10 are rejected under 35 U.S.C. 103 as being unpatentable over Yang in view of Simard, and further in view of Peraud et al. (US 2022/0230089).
With respect to claim 2, Yang in view of Simard teaches the computer-implemented method of claim 1, as discussed above. Yang in view of Simard doesn't expressly discuss wherein the corpus of documents is associated with a vector space.
Peraud teaches wherein the corpus of documents is associated with a vector space (Peraud, pa 0144, distances between the raw vectors and the seed vectors may be calculated. These may be Euclidean distances in a vector space.).
It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified with the teachings of Peraud because it represents the documents such that they can be analyzed for similarity (Peraud, pa 0145).
With respect to claim 5, Yang in view of Simard and Peraud teaches the computer-implemented method of claim 2, wherein evaluating classification performance of the prompt comprises: evaluating, via the one or more processors, the vector space to identify one or more clusters of documents (Peraud, pa 0144, distances between the raw vectors and the seed vectors may be calculated. These may be Euclidean distances in a vector space.).
With respect to claim 6, Yang in view of Simard and Peraud teaches the computer-implemented method of claim 5, wherein evaluating classification performance of the prompt further comprises: identifying, via the one or more processors, one or more respective clusters of documents in the vector space associated with one or more misclassifications of documents of the initial set of documents (Peraud, pa 0144, distances between the raw vectors and the seed vectors may be calculated. These may be Euclidean distances in a vector space. & pa 0159, collect all new verbatims 708 that were classified as an unknown topic, and these new verbatims 708 may be sent to the clustering model 716.).
With respect to claim 7, Yang in view of Simard and Peraud teaches the computer-implemented method of claim 6, wherein identifying that the initial set of documents does not include enough documents associated with the first issue of the one or more issues (Simard, pa 0086-0087, This guide 630 presents recommendations 632, 634 or recommended actions, suggestions, tips, or other information that guides the user in the process of refining the classification model to obtain better results. As shown in this example, a first recommendation 632 is presented along with a second recommendation 636. [0087] As illustrated by way of example in FIG. 6, recommendation #1 suggests “YOU DO NOT HAVE ENOUGH DOCUMENTS FOR A STATISTICALLY VALID SAMPLE TO OBTAIN YOUR DESIRED CONFIDENCE LEVEL.” A user interface element 634 is displayed to enable the user to remedy the deficiency by clicking on the recommended action (e.g., “CLICK HERE TO ADD MORE DOCUMENTS”).) comprises: evaluating, via the one or more processors, the one or more respective clusters of documents associated with the one or more misclassifications of documents and the corpus of documents to identify that the initial set of documents does not include enough documents from the one or more respective clusters of documents (Simard, pa 0079, A test run may be performed on a small training corpus of documents that the user selects for this purpose. Based on a review collection, the system displays classification metrics on a metric panel from this test run to provide the user with feedback on the accuracy of the model. The metrics displayed in the metrics panel enable the user to optimize a model's accuracy.).
With respect to claim 8, Yang in view of Simard and Peraud teaches the computer-implemented method of claim 6, wherein identifying that the corpus of documents is associated with the new issue comprises: evaluating, via the one or more processors, the one or more respective clusters of documents associated with the one or more misclassifications of documents and the prompt criteria to identify that at least one cluster of documents is not associated with at least one issue of the one or more issues (Peraud, pa 0158, if the classification model 714 did not assign the new verbatim 708 into any of the known topics, then the new verbatim 708 may be assigned to an unknown class… at inference time, new verbatims that are not assigned to the known topics can be set aside for new topic detection.).
With respect to claim 9, Yang in view of Simard and Peraud teaches the computer-implemented method of claim 5, wherein evaluating classification performance of the prompt further comprises: identifying, via the one or more processors, one or more respective clusters of documents in the vector space associated with one or more low-confidence classifications of documents of the initial set of documents (Peraud, pa 0159, The new verbatims 708 that were not classified into a known topic (i.e., were classified into the unknown topic category) may be processed by the clustering model 716).
With respect to claim 10, Yang in view of Simard and Peraud teaches the computer-implemented method of claim 9, wherein the one or more low- confidence classifications of documents are one or more of: weak classifications of documents, or documents with no classifications (Peraud, pa 0159, The new verbatims 708 that were not classified into a known topic (i.e., were classified into the unknown topic category) may be processed by the clustering model 716 & Lewis, Col. 13 Li. 55-62, Classification manager 205 ( e.g., using training module 210) selects 325 one or more classified documents 230 that are associated with a classification confidence level below a predetermined threshold value (e.g., 50%, 60%, or 75%) to create a set of low-confidence documents.).
Claims 3 and 4 are rejected under 35 U.S.C. 103 as being unpatentable over Yang in view of Simard and Peraud, and further in view of Lewis et al. (US 9,367,814).
With respect to claim 3, Yang in view of Simard and Peraud teaches the computer-implemented method of claim 2, as discussed above.
Lewis teaches wherein generating the request for documents comprises: identifying, via the one or more processors, one or more key terms associated with (1) the first issue of the one or more issues or (2) the new issue associated with the corpus of documents; and generating, via the one or more processors, the request for documents based on the one or more key terms (Lewis, Col. 10 Li. 3-12, Training module 210 associates document classifier A with label A and the attributes of label A. In particular, document classifier A is trained to recognize appropriate documents to associate with label A, namely documents including data that matches the attributes associated with label A. Training module 210 similarly associates document classifiers B and C with labels B and C, and the attributes of labels B and C, respectively. & Col. 11 Li. 27-38, classification manager 205 uses labels A, B, and C to generate document classifiers A, B, and C. In this example, classification manager 205 applies document classifiers A, B, and C to portions of data associated with corpus 110, including, e.g., unclassified documents. Based on an application of document classifiers 215 (e.g., document classifiers A, B, and C), classification manager 205 generates classified documents 230 by associating an unclassified Document 11 with label A, associating an unclassified Document 12 with label B, and associating an unclassified Document 13 with label B.).
It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified Yang in view of Simard and Peraud to have included the teachings of Lewis because the terms can indicate an appropriate topic of relevant information (Lewis, Col. 7 Li. 26-39).
With respect to claim 4, Yang in view of Simard, Peraud, and Lewis teaches the computer-implemented method of claim 3, wherein the one or more key terms are associated with one or more entities (Lewis, Col. 5 Li. 48-51, users of a product (illustrated as users 102, 103 and 104) provide information, such as complaints, comments relating to the product, etc. that forms at least a portion of the document corpus 110.).
Response to Arguments
35 U.S.C. 101
Applicant argues that claim provides an improvement by reciting “classifying, via the one or more processors, the obtained documents by inputting the prompt and the obtained documents into the generative Al model” and “evaluating, via the one or more processors, the classified obtained documents against ground truth data associated with the obtained documents to determine that the first issue or the new issue has been addressed.” The Examiner respectfully disagrees. The recited limitations are identified abstract ideas. At Step 2A Prong Two, the recited abstract idea should be integrated into a practical application through recitation of additional elements in the claim language. Here, there are not additional elements to provide such integration.
Applicant argues that the elements of amended claim 1 recite a specific technical improvement in how the prompt-based classification model operates providing a particular solution to a problem, similar to Ex Parte Desjardins. The Examiner respectfully disagrees. Ex Part Desjardins described claims that improved training of machine learning models. The present claims merely utilize generative AI models to perform tasks that can be done mentally. The limitations do not improve the training of a machine learning model or improve the computer performance. Instead, the computer is merely used as a tool to perform an existing process. See Enfish, LLC v. Microsoft Corp., 822 F.3d 1327, 1336, 118 USPQ2d 1684, 1689 (Fed. Cir. 2016).
Applicant argues that the amended claim 1 recites additional elements that amount to significantly more than the alleged abstract idea by providing an ordered combination of elements that is not generic by providing a specific iterative improvement to prompt-based classification models. The Examiner respectfully disagrees. Contrary to Appellant’s assertion, Appellant has not provided evidence (e.g., in the specification) for their assertion that elements are arranged in a non-conventional, non-generic arrangement of known, conventional elements. Viewing the limitations as an ordered combination does not add anything further than looking at the limitations individually. The recited limitations are abstract ideas and Applicant has not pointed to additional elements that amount to significantly more than the alleged abstract idea.
35 U.S.C. 103
Applicant seems to argue a newly amended limitation. Applicant’s amendment has rendered the previous rejection moot. Upon further consideration of the amendment, a new grounds of rejection is made in view of Yang et al. (US 12,579,179), and Simard et al. (US 2023/0315773).
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 BRITTANY N ALLEN whose telephone number is (571)270-3566. The examiner can normally be reached M-F 9 am - 5:00 pm EST.
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, Sherief Badawi can be reached at 571-272-9782. 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.
/BRITTANY N ALLEN/ Primary Examiner, Art Unit 2169
Prosecution Insights