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 .
DETAILED ACTION
This office action is in response to communication filed 4/4/2024. Claims 1-20 are currently pending and claims 1, 10, and 17 are the independent claims.
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.
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.
Claim 9 is 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.
As per claim 9, it recites “...wherein the first machine learning model is a simpler model than the second machine learning model.” The examiner would like to point out that different persons of ordinary skill in the art may have different opinions as to what would make a machine learning model/software/program/etc. “simpler”, ex: fewer operations performed on input, less computing resources consumed, organization of code structure, etc.., and as such the examiner is unclear as to what is meant by “simpler”. For the purpose of examination, the examination will consider these limitations to be “...wherein the first machine learning model consumes fewer resources than the second machine learning model.” as seen in the specification of this application in ex: pars. [0020], [0103], etc..
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.
As per claim 1, it recites “A system comprising: a memory that stores instructions; and one or more processors coupled to the memory and configured to execute the instructions to perform operations comprising: providing a support ticket for a software application to a trained machine learning model as input; receiving, from the trained machine learning model, a probability that the support ticket is addressed by modification of source code of the software application; based on the probability and a predetermined threshold, selecting a support group to send the support ticket to; and sending the support ticket to the selected support group. The limitation “based on the probability and a predetermined threshold, selecting a support group to send the support ticket to” as drafted, recites a function that, under its broadest reasonable interpretation, covers a function that could reasonably be performed in the mind, including with the aid of pen and paper, but for the recitation of generic computer components. As such, the limitation, as drafted, is a function that, under its broadest reasonable interpretation, recites the abstract idea of a mental process. The limitation encompasses a human mind carrying out the function through observation, evaluation, judgment, and /or opinion, or even with the aid of pen and paper. For example, a human may mentally judge/select/decide/etc. a group based on judged/observed/evaluated/analyzed/provided/etc. data/information/probability and threshold/etc.. Accordingly, this limitation recites and falls within the “Mental Processes” grouping of abstract ideas.
This judicial exception is not integrated into a practical application. The claim recites the following additional elements/limitations “A system comprising: a memory that stores instructions; and one or more processors coupled to the memory and configured to execute the instructions to perform operations comprising:”, “providing a support ticket for a software application to a trained machine learning model as input”, “receiving, from the trained machine learning model, a probability that the support ticket is addressed by modification of source code of the software application”, and “sending the support ticket to the selected support group.” The additional elements “A system comprising: a memory that stores instructions; and one or more processors coupled to the memory and configured to execute the instructions to perform operations comprising:” recite that high-level/generic computer/computer components are used to implement/perform the abstract idea/mental process and as such amounts to no more than mere instructions to apply the exception using generic computer, and/or mere computer components. The additional elements “providing a support ticket for a software application to a trained machine learning model as input”, “receiving, from the trained machine learning model, a probability that the support ticket is addressed by modification of source code of the software application”, and “sending the support ticket to the selected support group” do nothing more than add insignificant extra solution activities to the judicial exception of merely transmitting/gathering data/information and the courts have identified functions such as gathering, displaying, updating, transmitting and storing data as well-understood, routine, conventional activity, thus do not amount to significantly more than the judicial exception (see MPEP 2106.05(d)). Accordingly, the additional elements do not integrate the recited judicial exception into a practical application and the claim is therefore directed to the judicial exception. See MPEP 2106.05(f), 2106.05(g), etc..
The claim does 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 the exception using generic computer which is not significantly more than the abstract idea/mental process/judicial exception, and/or mere computer components, and mere insignificant extra solution activities to the judicial exception of merely transmitting/gathering data/information and the courts have identified functions such as gathering, displaying, updating, transmitting and storing data as well-understood, routine, conventional activity, thus do not amount to significantly more than the judicial exception (see MPEP 2106.05(d)). Accordingly, the claims are not patent eligible under 35 USC 101.
As per claim 2, it incorporates the deficiencies of claim 1, upon which it depends, and further recites “…wherein the operations further comprise: generating the trained machine learning model by providing a training set comprising a set of historical support tickets, each of the historical support tickets labeled with a class of a set of classes comprising a minority class and a majority class, more of the historical support tickets having majority class labels than minority class labels” which, conceptually, with broadest reasonable interpretation, further recites updating/modifying/training/ generating/etc. application/software/trained machine learning model/etc. used in implementing/performing/applying the abstract idea/mental process and insignificant extra solution activities and as such, with broadest reasonable interpretation, further recites an insignificant extra solution activities to the judicial exception of merely updating/modifying/etc. data/information/software/trained model and the courts have identified functions such as gathering, displaying, updating, transmitting and storing data as well-understood, routine, conventional activity, thus do not amount to significantly more than the judicial exception (see MPEP 2106.05(d)), and instructions to apply/implement/perform the abstract idea/judicial exception using generic computer/computer components/etc.. Therefore, the additional elements/limitations of claim 2 does not integrate the abstract idea into a practical application and is not significantly more than the abstract idea/mental process, and as such claim 2 is rejected for similar reasoning as claim 1, above.
As per claim 3, it incorporates the deficiencies of claim 1, upon which it depends, and further recites “…wherein the class that labels each historical support ticket identifies a support group that resolved the historical support ticket”, which, conceptually, with broadest reasonable interpretation, provides further as to the insignificant extra solution activities to the judicial exception of merely updating/modifying/etc. data/information/software/trained model and the courts have identified functions such as gathering, displaying, updating, transmitting and storing data as well-understood, routine, conventional activity, thus do not amount to significantly more than the judicial exception (see MPEP 2106.05(d)), and as such does not integrate the abstract idea into a practical application and is not significantly more than the abstract idea/mental process, and therefore claim 3 is rejected for similar reasoning as claim 1, above.
As per claim 4, it incorporates the deficiencies of claim 1, upon which it depends, and further recites “…wherein the operations further comprise: validating the trained machine learning model by determining a proportion of correct classifications of the historical support tickets classified by the trained machine learning model as the minority class out of a total number of classifications of the historical support tickets classified by the trained machine learning model as the minority class” which, conceptually, with broadest reasonable interpretation, provides further as to the abstract idea/mental process/judging/evaluating/validating/etc. performed, which does not integrate the abstract idea into a practical application and is not significantly more than the abstract idea/mental process. Therefore, claim 4 is rejected for similar reasoning as claim 1, above.
As per claim 5, it incorporates the deficiencies of claim 1, upon which it depends, and further recites “…wherein the operations further comprise: validating the trained machine learning model by determining a proportion of correct classifications of the historical support tickets classified by the trained machine learning model as the minority class out of a total number of the historical support tickets labeled as the minority class” which, conceptually, with broadest reasonable interpretation, provides further as to the abstract idea/mental process/judging/evaluating/validating/ determining/etc. performed, which does not integrate the abstract idea into a practical application and is not significantly more than the abstract idea/mental process. Therefore, claim 5 is rejected for similar reasoning as claim 1, above.
As per claim 6, it incorporates the deficiencies of claim 1, upon which it depends, and further recites “…wherein the minority class comprises no more than 10% of the support tickets and the majority class comprises at least 70% of the support tickets” which, conceptually, with broadest reasonable interpretation, provides further as to the abstract idea/mental process/judging/evaluating/validating/determining/etc. performed, which does not integrate the abstract idea into a practical application and is not significantly more than the abstract idea/mental process. Therefore, claim 6 is rejected for similar reasoning as claim 1, above.
As per claim 7, it incorporates the deficiencies of claim 1, upon which it depends, and further recites “…wherein the generating of the trained machine learning model comprises applying a balanced log loss function that penalizes misclassifications of the minority class more than misclassifications of the majority class” which, conceptually, with broadest reasonable interpretation, recites further clarification as to the insignificant extra solution activities of merely updating/modifying/training/generating/etc. data/information/software/trained model and the courts have identified functions such as gathering, displaying, updating, transmitting and storing data as well-understood, routine, conventional activity, thus do not amount to significantly more than the judicial exception (see MPEP 2106.05(d)). Therefore, the additional elements/limitations of claim 7 does not integrate the abstract idea into a practical application and are not significantly more than the abstract idea/mental process, and as such claim 7 is rejected for similar reasoning as claim 1, above.
As per claim 8, it incorporates the deficiencies of claim 1, upon which it depends, and further recites “…wherein: the trained machine learning model is a second machine learning model; the probability that the support ticket is addressed by modification of source code of the software application is a second probability; the predetermined threshold is a second predetermined threshold; the operations further comprise: providing the support ticket for a software application to a first trained machine learning model as input; and receiving, from the first trained machine learning model, a first probability that the support ticket is addressed by modification of source code of the software application; and the providing of the support ticket for the software application to the second machine learning model is based on the first probability and a first predetermined threshold” which, conceptually, with broadest reasonable interpretation, provides further clarification as to the abstract idea/mental process and insignificant extra solution activities performed, and recites further insignificant extra solution activities of transmitting/gathering data/information/support ticket/probability/etc., and the courts have identified functions such as gathering, displaying, updating, transmitting and storing data as well-understood, routine, conventional activity, thus do not amount to significantly more than the judicial exception (see MPEP 2106.05(d)). Therefore, the additional elements/limitations of claim 8 do not integrate the abstract idea into a practical application and are not significantly more than the abstract idea/mental process, and as such claim 8 is rejected for similar reasoning as claim 1, above.
As per claim 9, it incorporates the deficiencies of claim 1, upon which it depends, and further recites “…wherein the first machine learning model is a simpler model than the second machine learning model” which, conceptually, with broadest reasonable interpretation, provides further clarification as to the computer/computer components/software/machine learning models/etc. used to implement/perform the abstract idea/mental process/extra solution activities/etc., which does not integrate the abstract idea into a practical application and is not significantly more than the abstract idea/mental process. Therefore, claim 9 is rejected for similar reasoning as claim 1, above.
As per claim 10, it recites a non-transitory computer-readable medium having similar limitations as the system of claim 1, and as such recites a similar abstract idea and has similar deficiencies as claim 1. Claim 10 recites the further elements/limitations “A non-transitory computer-readable medium that stores instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising” which, conceptually, with broadest reasonable interpretation, recites high level/generic computer/computer components used to implement/perform/apply the abstract idea/mental process, which does not integrate the abstract idea into a practical application and is not significantly more than the abstract idea/mental process. As such, the additional limitations/elements of claim 10 fail to correct the deficiencies of claim 1, and therefore claim 10 is rejected for similar reasoning as claim 1, above.
As per claims 11-16, they recite non-transitory computer-readable mediums having similar limitations as the systems of claims 2-7, respectively, and are therefore rejected for similar reasoning as claims 2-7, respectively, above.
As per claim 17, it recites a method having similar limitations as the operations performed by the system of claim 1, and as such recites a similar abstract idea and has similar deficiencies as claim 1. Therefore claim 17 is rejected for similar reasoning as claim 1, above.
As per claims 18-20, they recite methods having similar limitations as the systems of claims 2-4, respectively, and are therefore rejected for similar reasoning as claims 2-4, respectively, above.
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, 8, 10, and 17 are rejected under 35 U.S.C. 103 as being unpatentable over MacLaughlin et al. (herein called MacLaughlin) (US PG Pub. 2025/0165882 A1) and Burton et al. (herein called Burton) (US PG Pub. 2019/0026697 A1).
As per claim 1, MacLaughlin teaches: a system comprising: a memory that stores instructions; and one or more processors coupled to the memory and configured to execute the instructions to perform operations comprising:
providing a support ticket for a software application to a trained machine learning model as input (pars. [0003], [0029], [0039]-[0040], [0052], work ticket including request for assistance with a software issue is input to/received by/etc. artificial intelligence model/machine learning model/group mapping model/agent mapping model/etc. (providing a support ticket for a software application to a trained machine learning model as input) that determines group/agent to assign work ticket/support ticket to resolve the issue.);
receiving, from the trained machine learning model, a probability that the support ticket is addressed by the software application (pars. [0003], [0022]-[0023], [0039]-[0047], [0050], [0055], [0058], artificial intelligence/trained model/ticketing models/group mapping model/agent mapping model/etc. predicts/determines/etc. work ticket to be a software issue based on description of issue in work ticket and predicts a software agent group/agent in a software agent group/etc. having expertise needed to resolve the issue/best able to handle the issue/available to resolve the issue/etc., and assigns the work ticket to the software agent group/agent/etc.. As the trained models predict that the issue in the work ticket is a software issue and predicts a software agent group/software agent having expertise/best suited/available/etc. to assign the ticket to resolve the software issue, the prediction that the issue in the work ticket is a software issue and of the group/agent best suited to resolve the software issue is a probability that the support/work ticket is addressed/resolved/etc. by/through/etc. the software application/software/etc.);
based on the probability and a predetermined threshold, selecting a support group to send the support ticket to (pars. [0022]-[0023], [0039]-[0044], [0050], [0055], [0058], artificial intelligence/trained model/ticketing models/group mapping model/agent mapping model/etc. predicts/determines/etc. work ticket to be a software issue based on description of issue in work ticket and predicts a software agent group/agent in a software agent group/etc. having expertise needed to resolve the issue/best able to handle the issue/available to resolve the issue/etc. (probability support ticket is addressed/resolved by software application), determines maximum workload/maximum work distribution (predetermined threshold) of software agents/software agents groups resolving issues/work tickets, and predicts agent/software agent group/etc. to assign the work ticket to the software agent group/agent/etc. based on the agent group/agent being predicted to have the expertise to resolve the issue/best suited to resolve issue/etc. and based on the maximum workload/maximum work distribution/predetermined threshold of work tickets/etc. of the agent groups/agents/support groups (select support group/software agent group/agent/etc. to send/assign/route/etc. the support ticket/work ticket based on the probability/predicted agent group/agents/etc. and a predetermined threshold/max workload/max work distribution/etc.).); and
sending the support ticket to the selected support group (pars. [0003], [0022]-[0023], [0059], work ticket/support ticket is routed/assigned/etc. to predicted/selected/etc. team/group/agent (sending/routing/assigning/etc. support/work ticket to selected support group/team/group/agent/etc.) who works to resolve the issues/ticket/etc..).
While MacLaughlin teaches that a machine learning model routes/directs/assigns/etc. work tickets associated with software issues to groups/agents/developers/etc. that are capable of resolving the software issue, it does not explicitly state that resolving the software issue includes modifying the software, and as such does not explicitly state, however Burton teaches:
the support ticket is addressed by modification of source code of the software application (pars. [0006], [0018], [0021]-[0022], [0028], [0032], software has source code which is maintained/changed/updated/etc. by developers, and software issue reports (support ticket) are used to create workflow/tasks/etc. which are assigned to developers/users/groups/etc. who then make code changes/source code changes and submit pull requests for the source code to correct/fix/address the software issue reports (support ticket/software issue report is addressed/issue is fixed/etc. by assigned developer/user/group making source code changes/modification of source code of the software application, and as MacLaughlin teaches that the ticket may be for a software issue and is routed/assigned/etc. to group/agent/developer predicted/determined to be probable/etc. to be able to handle/correct/fix/etc. the software issue in the ticket, it is obvious that the support ticket is addressed/resolved/handled/etc. by modification of source code of the software application.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to add the support ticket is addressed by modification of source code of the software application, as conceptually taught by Burton, into that of MacLaughlin because these modifications allow for developers/users/etc. to modify/update/correct/fix/etc. source code of software as needed/desired in order to correct/fix/etc. issues/bugs/problems/etc. in the software, which is desirable as it allows for errors/bugs/issues to be corrected so that the software operates correctly/as desired.
As per claim 8, MacLaughlin further teaches: wherein: the trained machine learning model is a second machine learning model; the probability that the support ticket is addressed by modification of source code of the software application is a second probability; the predetermined threshold is a second predetermined threshold (fig. 5 items 504 and 510, pars. [0003], [0007], [0023], [0039]-[0045], [0050], [0055]-[0058], multiple machine learning models may be used to assign work ticket to agent including a group ticketing/mapping model (first trained machine learning model) that analyzes the ticket and predicts that it is a software issue that is best resolved by a software agent group and assigns the ticket to the agent group based on expertise/experience/etc. of agent group, minimum/maximum description length of software issue in ticket/thresholds/etc. and sends the ticket to an agent ticketing/mapping model (second trained machine learning model) which analyzes the ticket and predicts an agent to assign the ticket to based on expertise/experience/etc. of agent and maximum workload/maximum workload distribution/thresholds/etc.. As there are two machine learning models/group ticketing model and agent ticketing model that predict/determine probability/etc. group and agent to assign software issue ticket to, it is obvious that the trained machine learning model may be the agent ticketing module/a second machine learning model and that the prediction and assignment of the ticket to the agent using maximum workload/threshold may be a second probability that the support ticket is addressed by modification of source code and the maximum workload is a second predetermined threshold as the prediction and assignment made by the group ticketing model would be a first probability using first threshold.);
the operations further comprise: providing the support ticket for a software application to a first trained machine learning model as input (pars. [0025], [0039]-[0041], [0050], work ticket for software issue/support ticket for a software application is received by group ticketing/mapping model (provide support ticket to first machine learning model as input)); and
receiving, from the first trained machine learning model, a first probability that the support ticket is addressed by modification of source code of the software application (pars. [0003], [0039]-[0041], [0050], [0055], group mapping/ticketing model/first trained machine learning model evaluates description of issue is work ticket and predicts that it is a software issue and predicts a software agent group to assign the ticket to for resolution of the issue. As the group model/first model provides a prediction that the issue in the work ticket is a software issue and predicts an agent group to resolve the software issue, and Burton teaches that resolving ticket issues includes modifying software as seen in the rejection of claim 1 above, it is obvious the prediction is a first probability that the work ticket/support ticket is addressed by modification of source code of the application.); and
the providing of the support ticket for the software application to the second machine learning model is based on the first probability and a first predetermined threshold (pars. [0039]-[0041], [0050], [0058], group mapping/ticketing model/first machine learning model evaluates/analyzes/etc. description of issue in work ticket that has minimum/maximum description length (first predetermined threshold) and predicts that the issue is a software issue and predicts a software agent group to resolve the issue (first probability/prediction), and work ticket/support ticket, predicted agent group data, and agent data is routed/provided/etc. to agent mapping/ticketing model/second machine learning model (providing of the support ticket for the software application to the second machine learning model) which predicts an agent to assign the ticket to for resolution of the issue. As the agent mapping model/second machine learning model receives and processes the work ticket/agent group data/agent data/etc. to predict an agent to assign the ticket to after the group mapping model/first model predicts the agent group to assign the ticket having the predicted software issue to and the group mapping/ticketing model makes its predictions/first probability using a description of the software issue having a minimum/maximum length/first predetermined threshold, it is obvious that the providing of the support ticket for the software application to the second machine learning model is based on the first probability and a first predetermined threshold.).
As per claims 10 and 17 they recite a non-transitory computer readable medium and a method, respectively, having similar limitations as the system of claim 1, and are therefore rejected for similar reasoning as claim 1, above.
Claims 2-3, 11-12, and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over MacLaughlin et al. (herein called MacLaughlin) (US PG Pub. 2025/0165882 A1) and Burton et al. (herein called Burton) (US PG Pub. 2019/0026697 A1) in further view of Zhang (US Patent 11,182,691 B1).
As per claim 2, MacLaughlin further teaches: wherein the operations further comprise: generating the trained machine learning model by providing a training set comprising a set of historical support tickets (pars. [0025], [0040], [0042], [0046], [0049], machine learning models/agent mapping model/group mapping model/etc. are trained (generate trained machine learning model) using training dataset including historical data such as historical ticket data (provide training set comprising set of historical support tickets).).
MacLaughlin does not explicitly state, however teaches: each of the historical support tickets labeled with a class of a set of classes comprising a minority class and a majority class, more of the historical support tickets having majority class labels than minority class labels (col. 54 lines 7-58, col. 55 lines 38-64, col. 60 lines 31-36, 59-63, col. 63 lines 4-7, training set/records/training data/etc. used to train machine learning model (historical support tickets from MacLaughlin) are classified into majority and minority categories (labeled with a class of set of classes comprising a minority and majority class).).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to add each of the historical support tickets labeled with a class of a set of classes comprising a minority class and a majority class, more of the historical support tickets having majority class labels than minority class labels, as conceptually taught by Zhang, into that of MacLaughlin and Burton because these modifications allow for the data/information/tickets/etc. used to train/generate/etc. the models to be classified/categorized/etc. into different types/classes/categories thereby helping to ensure that data from each of the classes/types/categories/etc. to be used to train/generate the model, which is desirable as it helps ensure that the model can successfully process/classify/etc. different desired types/classes/etc. of data/information increasing the usability of the model while helping to ensure the model operates correctly/as desired.
As per claim 3, MacLaughlin further teaches: wherein the class that labels each historical support ticket identifies a support group that resolved the historical support ticket (figs. 3A and 3B, pars. and as seen if figs 3A and 3B the tickets/historical support tickets include ticket data/labels/class labels/etc. that includes the assignment group, agent ticket is assigned to, and agent ticket is resolved by (tickets/historical support tickets is labeled identifying support group that resolved the ticket/historical support ticket).).
As per claims 11-12 they recite non-transitory computer readable mediums having similar limitations as the systems of claims 2-3, respectively, and are therefore rejected for similar reasoning as claims 2-3 respectively, above.
As per claims 18-19 they recite methods having similar limitations as the systems of claims 2-3, respectively, and are therefore rejected for similar reasoning as claims 2-3 respectively, above.
Claims 4-7, 13-16, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over MacLaughlin et al. (herein called MacLaughlin) (US PG Pub. 2025/0165882 A1), Burton et al. (herein called Burton) (US PG Pub. 2019/0026697 A1), and Zhang US Patent 11,182,691 B1), in further view of Healy et al. (herein called Healy) (US PG Pub. 2021/0406369 A1).
As per claim 4, MacLaughlin, Burton, and Zhang do not explicitly state, however Healy teaches:
wherein the operations further comprise: validating the trained machine learning model by determining a proportion of correct classifications of the historical support tickets classified by the trained machine learning model as the minority class out of a total number of classifications of the historical support tickets classified by the trained machine learning model as the minority class (pars. [0011], [0026], [0038], [0055], [0058], machine learning model classifies data/files/records/etc. (historical support tickets from MacLaughlin) into classes including minority class with a desired/target/etc. amount of accuracy/80%, 95%, 99%, etc. (proportion of correct classification), and resulting classifications data/training data/records/support tickets/etc. (tickets classified by model as minority class) are compared to their known/labeled/actual/etc. classification to determine machine learning accuracy/ratio and percentage of misclassifications and correct classification/etc. and determines if it satisfies a threshold/meets target accuracy/etc. (determine proportion/ratio/percentage/etc. of correct classifications of the historical support tickets classified by the trained machine learning model as the minority class out of a total number of classifications of the historical support tickets classified by the trained machine learning model as the minority class to validate the model/determine if accuracy meets threshold/etc.).).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to add wherein the operations further comprise: validating the trained machine learning model by determining a proportion of correct classifications of the historical support tickets classified by the trained machine learning model as the minority class out of a total number of classifications of the historical support tickets classified by the trained machine learning model as the minority class, as conceptually taught by Healy, into that of MacLaughlin, Burton, and Zhang because these modifications allow for the machine learning model to be checked/validated/etc. to make sure it is operating correctly/correctly classifying data/etc., thereby helping to prevent errors and ensure that the models are operating as desired.
As per claim 5, MacLaughlin, Burton, and Zhang do not explicitly state, however Healy teaches:
wherein the operations further comprise: validating the trained machine learning model by determining a proportion of correct classifications of the historical support tickets classified by the trained machine learning model as the minority class out of a total number of the historical support tickets labeled as the minority class (pars. [0011], [0026], [0038], [0055], [0058], machine learning model classifies data/files/records/etc. (historical support tickets from MacLaughlin) into classes including minority class with a desired/target/etc. amount of accuracy/80%, 95%, 99%, etc. (proportion of correct classification), and resulting classifications data/training data/records/support tickets/etc. (tickets classified by model as minority class) are compared to their known/labeled/actual/etc. classification to determine machine learning accuracy/ratio and percentage of misclassifications and correct classification/etc. and determines if it satisfies a threshold/meets target accuracy/etc. (determine proportion/ratio/percentage/etc. of correct classifications of the historical support tickets classified by the trained machine learning model as the minority class out of a total number of the historical support tickets labeled as the minority class to validate the model/determine if accuracy meets threshold/etc.).).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to add wherein the operations further comprise: validating the trained machine learning model by determining a proportion of correct classifications of the historical support tickets classified by the trained machine learning model as the minority class out of a total number of the historical support tickets labeled as the minority class, as conceptually taught by Healy, into that of MacLaughlin, Burton, and Zhang because these modifications allow for the machine learning model to be checked/validated/etc. to make sure it is operating correctly/correctly classifying data/etc., thereby helping to prevent errors and ensure that the models are operating as desired.
As per claim 6, MacLaughlin and Burton do not explicitly state, however Zhang teaches:
wherein the minority class comprises no more than 10% of the support tickets and the majority class comprises at least 70% of the support tickets (col. 54 lines 7-58, col. 55 lines 38-64, col. 60 lines 31-36, 59-63, records/data/etc. (support tickets from MacLaughlin) classified in minority category may be rare/unusual with occurrence happening once per thousand/hundred thousand/million/etc., have a majority/minority population ration of 100:1/1000:1/etc., etc.. As the ratio of majority to minority classification of records/support tickets may be hundred/thousand/million/etc. majority to 1 minority, it is obvious that the majority class comprises over 70% of the support tickets/records and the minority class comprises less than/no more than/etc. 10% of the records/support tickets.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to add e wherein the minority class comprises no more than 10% of the support tickets and the majority class comprises at least 70% of the support tickets, as conceptually taught by Zhang, into that of MacLaughlin, Burton, and Healy because these modifications allow for the data/information/tickets/etc. used to train/generate/etc. the models to be classified/categorized/etc. into different types/classes/categories thereby helping to ensure that data from each of the classes/types/categories/etc. to be used to train/generate the model, which is desirable as it helps ensure that the model can successfully process/classify/etc. different desired types/classes/etc. of data/information increasing the usability of the model while helping to ensure the model operates correctly/as desired.
As per claim 7, MacLaughlin and Burton do not explicitly state, however Zhang teaches:
wherein the generating of the trained machine learning model comprises applying a balanced log loss function that penalizes misclassifications of the minority class more than misclassifications of the majority class (col. 54 lines 15-36, 60-67, col. 56 lines 16-52, machine learning performance goals are set including classification accuracy of classifying data into categories/majority and minority categories/etc., and when accuracy of classification of minority may be low (misclassifications of minority class) sampling ratio of training data is used to select training data to train machine learning model such that higher percentages of training data classified as minority and less data classified as majority is used to generate/train machine learning model (generating trained machine learning model) resulting in higher prediction accuracy for minority categories (applying a balanced log loss function that penalizes misclassifications of the minority class more than misclassifications of the majority class/apply sampling ratio to training data when accuracy of minority classifications is low to increase prediction accuracy of minority categories while using less training data classified in majority category).).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to add wherein the generating of the trained machine learning model comprises applying a balanced log loss function that penalizes misclassifications of the minority class more than misclassifications of the majority class, as conceptually taught by Zhang, into that of MacLaughlin, Burton, and Healy because these modifications allow for the machine learning model to be trained to correctly classify data into minority categories thereby helping to ensure that they may accurately classify the data and operate correctly/as desired by users.
As per claims 13-16 they recite non-transitory computer readable mediums having similar limitations as the systems of claims 4-7, respectively, and are therefore rejected for similar reasoning as claims 4-7 respectively, above.
As per claim 20, it recites a method having similar limitations as the system of claim 4, and is therefore rejected for similar reasoning as claim 4, above.
Claims 9 are rejected under 35 U.S.C. 103 as being unpatentable over MacLaughlin et al. (herein called MacLaughlin) (US PG Pub. 2025/0165882 A1) and Burton et al. (herein called Burton) (US PG Pub. 2019/0026697 A1) in further view of Healy et al. (herein called Healy) (US PG Pub. 2021/0406369 A1).
As per claim 9, while MacLaughlin teaches using a first and second machine learning model/different machine learning models/etc., MacLaughlin and Burton do not explicitly state, however Healy teaches:
wherein the first machine learning model is a simpler model than the second machine learning model (pars. [0011]-[0012], [0028], different machine learning models may have different amounts of complexity, may consume different amounts of computing resources/memory/processor resources/etc., etc. such that one model may consume fewer/less/etc. resources than another model/be less complex than another model/etc.. As different models/machine learning models may consume different amounts of resources/have different complexity levels/etc., it is obvious that the first machine learning model may be a simpler/less complex/consumer fewer resources/etc. than the second machine learning model.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to add wherein the first machine learning model is a simpler model than the second machine learning model, as conceptually taught by Healy, into that of MacLaughlin and Burton, because these modifications allow for different/multiple/etc. machine learning models to be used that may have different complexities/consume different amounts of resources/etc., which is desirable as it increases usability by allowing for more/different/etc. types of machine learning models to be used thereby increasing user control over the machine learning models used and helping to ensure that the models operate as desired by users/correctly/etc..
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Mahanta et al. US Patent 11,726,775 B2 teaches that machine learning models may be used to determine candidate developers/users/etc. to correct identified issues in source code.
Gaddam et al. US PG Pub. 2025/0130775 A1 teaches that issues/tasks/jobs/etc. requiring code modifications may be determined and developers/individuals/etc. suited to address the issue/make the code modifications/associated with the job/code/etc. are notified so they may make the code changes.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DOUGLAS M SLACHTA whose telephone number is (571)270-0653. The examiner can normally be reached Monday-Friday 6:30am-4pm.
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, Chat Do can be reached at 571-272-3721. 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.
/DOUGLAS M SLACHTA/Examiner, Art Unit 2193
Prosecution Insights