CTNF 18/401,813 CTNF 87075 DETAILED ACTION Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Information Disclosure Statement The information disclosure statements (IDS) submitted on May 28, 2026 are being considered by the examiner. Double Patenting 08-33 AIA The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg , 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman , 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi , 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum , 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel , 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington , 418 F.2d 528, 163 USPQ 644 (CCPA 1969). A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA. A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b). The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13. The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA/25, or PTO/AIA/26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer. Claims 1-20 are provisionally rejected on the ground of nonstatutory double patenting as being unpatentable over the claims of copending Application No. 18/536,422 18/401,810, 18/401,807, and 18/401,805 (reference applications). Although the claims at issue are not identical, they are not patentably distinct from each other because the claims of the instant application are obvious variants of the claims of the co-pending applications. It is noted that through prosecution of all related cases, the double patenting rejection may change based on amendments through the course of examination of the instant and reference applications. This is a provisional nonstatutory double patenting rejection because the patentably indistinct claims have not in fact been patented. Claim Rejections - 35 USC § 102 07-07-aia AIA 07-07 The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action: A person shall be entitled to a patent unless – 07-08-aia AIA (a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention. 07-12-aia AIA (a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention. 07-15 AIA Claim s 8-14 are rejected under 35 U.S.C. 102( a)(1)/(a)(2 ) as being anticipated by Iyer et al. (U.S. Patent No. 11,836,163 B1, hereinafter referred to as “Iyer”) . Regarding claim 8, Iyer discloses a computing system facilitating data drift detection, comprising: at least one processor; a communication interface communicatively coupled to the at least one processor; and a memory device storing executable code that, when executed, causes the at least one processor to: (“The draft drift detection compute device 110 includes a processor 112 operatively coupled to a memory 114 (e.g., via a system bus). The data drift detection compute device 110 can be any type of device, such as a server, desktop, laptop, mobile device, internet of things device, and/or the like.”)(e.g., figure 1 and col 4 lines 16-22 and col 5 lines 63-67) perform data processing on one or more datasets; (“For example, the machine learning model 156 may be used to perform functions related to natural language processing (e.g., analyzing speech, analyzing entities, analyzing sentiment, etc.) or computer vision (e.g., analyzing images, analyzing video, etc.). The machine learning model 156 may be designed and/or trained at any compute device, such as data drift detection compute device 110, user compute device 150, and/or a compute device not shown in FIG. 1.” “the machine learning model 156 is associated with a natural language processing task, and the sets of vector representations 118A and 118B are embedding vectors. Using embedding vectors allows semantic similarity of associated natural langue processing data to be quantified (e.g., by how close the vectors are in vector space).”)(e.g., col 5 lines 28-36 and col 9 lines 55-61) derive, from the one or more datasets, data features that would be used in data analysis; (“Some techniques described herein generate and use vector representations of higher dimension data and/or unstructured data, such as certain forms of images, images with captions, free form text, multi-modal data, video, audio, social media posts, and/or the like. In some instances, generating vector representations from higher dimension data and/or unstructured data can enable data drift to be detected with respect to that higher dimension data and/or unstructured data, something that may not have been possible if vector representations were not generated.”)(e.g., col 3 lines 65 – col 4 line 7) classify the data features as being textual data features; (“Additionally, in some instances, vector representations can account for semantic similarity, thereby allowing similarity and/or dissimilarity in meaning between text to be identified.” – textual data features)(e.g., col 18 lines 47-50). compare text lens numerical values and text sentiment numerical values of incoming data relative historical data; (“For example, the machine learning model 156 may be used to perform functions related to natural language processing (e.g., analyzing speech, analyzing entities, analyzing sentiment, etc.)…” “In some implementations, the number of clusters to be included in the set of clusters 120 can be a function of a size (e.g., count) of the set of vector representations 118A and/or set of data 116A. For example, as the number of vector representations included in the set of vector representations 118A increases, the number of clusters included in the set of clusters 120 can also increase.” “A first histogram is shown at chart 206, where N can represent, for example, a range of count values, a range of normalized count values, a range of probability values, and/or the like.”)(e.g., col 5 lines 28-32, col 10 lines 30-36 and col 11 lines 35-38) apply a statistical test to textual data features of the incoming data and the historical data to determine whether a statistically significant change exists between the incoming data and the historical data, (“In some implementations, distributions for those additional sets of vector representations can be determined based on the set of clusters 120 and set of statistical properties 122. In some implementations, distributions for those additional sets of vector representations can be determined based on a different set of clusters and statistical properties, such as a set of clusters and associated statistical properties determined based on the set of vector representations 118B and/or a set of vector representations not shown in FIG. 1. As such, the set of clusters and set of statistical properties used for generating the distribution that is to act as the baseline and be compared against another distribution(s) (e.g., from production) for detecting data drift can change over time.”)(e.g., col 9 lines 41-54) the statistical test incorporating a population stability index score; and (“the set of divergence metrics are detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback-Leibler divergence.” – Kullback-Leibler is considered to incorporate a population stability index score)(e.g., col 19 lines 29-32) determine that one or more statistically significant changes exist causing data drift. (“At 409, data drift is detected between the first set of vector representations and the second set of vector representations based on a comparison of the distribution associated with the first set of vector representations with the distribution associated with the second set of vector representations. In some implementations, the data drift is detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback-Leibler divergence. In some implementations, step 409 is performed automatically (e.g., without requiring additional human input) in response to completing step 408.” “At 506, data drift between the first set of vector representations and the second set of vector representations is detected based on a comparison of the distribution associated with the first set of vector representations with the distribution associated with the second set of vector representations at 505. In some implementations, step 506 is performed automatically (e.g., without requiring additional human input) in response to completing step 505.”)(e.g., figures 4B and 5 and col 14 lines 20-30 and col 17 lines 10-17) Regarding claim 9, Iyer discloses the computing system of claim 8. Iyer further discloses wherein the determining that the one or more statistically significant changes exist is based on the population stability index score surpassing a threshold value. (“At 409, data drift is detected between the first set of vector representations and the second set of vector representations based on a comparison of the distribution associated with the first set of vector representations with the distribution associated with the second set of vector representations. In some implementations, the data drift is detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback-Leibler divergence. In some implementations, step 409 is performed automatically (e.g., without requiring additional human input) in response to completing step 408.” “At 410, transmission of a signal is caused, in response to the data drift exceeding a data drift threshold, to cause a remedial action. With reference to FIG. 1, the signal could be transmitted to the data drift detection device 110, user compute device 150, and/or a compute device not shown in FIG. 1.” “At 506, data drift between the first set of vector representations and the second set of vector representations is detected based on a comparison of the distribution associated with the first set of vector representations with the distribution associated with the second set of vector representations at 505. In some implementations, step 506 is performed automatically (e.g., without requiring additional human input) in response to completing step 505.” “At 507, transmission of a signal is caused, in response to the data drift exceeding a data drift threshold, to cause a remedial action. With reference to FIG. 1, the signal could be transmitted to the data drift detection device 110, user compute device 150, and/or a compute device not shown in FIG. 1.”)(e.g., figures 4B and 5 and col 14 lines 20-30 and 31-36 and col 17 lines 10-17 and 18-23). Regarding claim 10, Iyer discloses the computing system of claim 9. Iyer further discloses wherein the executable code, when executed, further causes the at least one processor to set the threshold value. (“If data drift is considered to have occurred and/or an amount of the draft drift exceeds a predetermined data drift threshold, a remedial action can occur and/or can be initiated (triggered). For example, the data drift detection compute device 110 can send a signal to user compute device 150 and/or a compute device not shown in FIG. 1, alerting the presence of data drift and/or data drift beyond the predetermined data drift threshold.” - system includes threshold that has been set)(e.g., col 8 lines 22-29). Regarding claim 11, Iyer discloses the computing system of claim 8. Iyer further discloses wherein the executable code, when executed further causes the at least one processor to perform natural language processing on text, and (“the machine learning model is a natural language processing machine learning model.”)(e.g., col 17 lines 28-29) based thereon assign the text lens numerical values and the text sentiment numerical values. (“For example, the machine learning model 156 may be used to perform functions related to natural language processing (e.g., analyzing speech, analyzing entities, analyzing sentiment, etc.)…” “In some implementations, the number of clusters to be included in the set of clusters 120 can be a function of a size (e.g., count) of the set of vector representations 118A and/or set of data 116A. For example, as the number of vector representations included in the set of vector representations 118A increases, the number of clusters included in the set of clusters 120 can also increase.” “A first histogram is shown at chart 206, where N can represent, for example, a range of count values, a range of normalized count values, a range of probability values, and/or the like.”)(e.g., col 5 lines 28-32, col 10 lines 30-36 and col 11 lines 35-38). Regarding claim 12, Iyer discloses the computing system of claim 8. Iyer further discloses wherein the executable code, when executed further causes the at least one processor to, based on determining that at least one machine learning model relies upon the data features, identify one or more administrative users that oversees management of the at least one machine learning model. (“For example, the remedial action could include causing an alert indicating that the data drift exceeds the data drift threshold to be triggered. As another example, the remedial action could include causing a compute device (e.g., data drift detection compute device 110 and/or user compute device 150) and/or machine learning model (e.g., machine learning model 156) to change and/or modify a mode of operation, such as refraining from producing additional output data, refraining from relying on recently produced output data, logging events, shutting down, operating in a restricted access mode, and/or the like. As another example, the remedial action could include causing a machine learning model using the second set of data to be retrained.” – users that are notified are considered to be administrative users that can implement the remedial action.)(e.g., col 14 lines 36-49). Regarding claim 13, Iyer discloses the computing system of claim 12. Iyer further discloses wherein the executable code, when executed further causes the at least one processor to transmit an electronic notification to respective computing devices associated with the one or more administrative users. (“For example, the data drift detection compute device 110 can send a signal to user compute device 150 and/or a compute device not shown in FIG. 1, alerting the presence of data drift and/or data drift beyond the predetermined data drift threshold.” “For example, the remedial action could include causing an alert indicating that the data drift exceeds the data drift threshold to be triggered.”)(e.g., col 8 lines 25-29 and col 14 lines 36-38). Regarding claim 14, Iyer discloses the computing system of claim 13. Iyer further discloses receive, from a computing device of the respective computing devices, a request to retrain the at least one machine learning model with the incoming data. (“As another example, the remedial action could include causing a machine learning model using the second set of data to be retrained.”)(e.g., col 14 lines 47-49) . Claim Rejections - 35 USC § 103 07-20-aia AIA 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. 07-23-aia AIA The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows: 1. Determining the scope and contents of the prior art. 2. Ascertaining the differences between the prior art and the claims at issue. 3. Resolving the level of ordinary skill in the pertinent art. 4. Considering objective evidence present in the application indicating obviousness or nonobviousness. 07-20-02-aia AIA This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention. 07-21-aia AIA Claim s 1, 2, 4-7, 15, 16 and 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over Iyer in view of Bareliyahu et al. (U.S. Publication No. 2024/0362360 A1, hereinafter referred to as “Bareliyahu”) . Regarding claim 1, Iyer discloses a computing system facilitating database and data structure management, comprising: at least one processor; a communication interface communicatively coupled to the at least one processor; and a memory device storing executable code that, when executed, causes the at least one processor to: (“The draft drift detection compute device 110 includes a processor 112 operatively coupled to a memory 114 (e.g., via a system bus). The data drift detection compute device 110 can be any type of device, such as a server, desktop, laptop, mobile device, internet of things device, and/or the like.”)(e.g., figure 1 and col 4 lines 16-22 and col 5 lines 63-67) perform data processing on one or more datasets; (“For example, the machine learning model 156 may be used to perform functions related to natural language processing (e.g., analyzing speech, analyzing entities, analyzing sentiment, etc.) or computer vision (e.g., analyzing images, analyzing video, etc.). The machine learning model 156 may be designed and/or trained at any compute device, such as data drift detection compute device 110, user compute device 150, and/or a compute device not shown in FIG. 1.” “the machine learning model 156 is associated with a natural language processing task, and the sets of vector representations 118A and 118B are embedding vectors. Using embedding vectors allows semantic similarity of associated natural langue processing data to be quantified (e.g., by how close the vectors are in vector space).”)(e.g., col 5 lines 28-36 and col 9 lines 55-61) derive, from the one or more datasets, data features that would be used in data analysis; (“Some techniques described herein generate and use vector representations of higher dimension data and/or unstructured data, such as certain forms of images, images with captions, free form text, multi-modal data, video, audio, social media posts, and/or the like. In some instances, generating vector representations from higher dimension data and/or unstructured data can enable data drift to be detected with respect to that higher dimension data and/or unstructured data, something that may not have been possible if vector representations were not generated.”)(e.g., col 3 lines 65 – col 4 line 7) classify the data features as either being categorical or numerical; (“In some implementations, each vector representation from the set of vector representations 118A is associated with (e.g., is linked to, is binned to) a cluster from the set of clusters 120 that is most similar to (e.g., closest distance in vector space) that vector representation. In some instances, multiple different vector representations from the set of vector representations 118A can be associated with a common cluster from the set of clusters 120. In some implementations, the distribution 124A can indicate, for each vector representation from the set of vector representations 118A, a cluster from the set of clusters 120 whose associated statistical property from the set of statistical properties 122 to which that vector representation is most similar. As such, the distribution 124A can also indicate the number of vector representations from the set of vector representations 118A associated with a given cluster from the set of clusters 120.” “In some implementations, each vector representation from the set of vector representations 118A is associated with (e.g., is linked to, is binned to) a cluster from the set of clusters 120 that is most similar to (e.g., closest distance in vector space) that vector representation. In some instances, multiple different vector representations from the set of vector representations 118A can be associated with a common cluster from the set of clusters 120. In some implementations, the distribution 124A can indicate, for each vector representation from the set of vector representations 118A, a cluster from the set of clusters 120 whose associated statistical property from the set of statistical properties 122 to which that vector representation is most similar. As such, the distribution 124A can also indicate the number of vector representations from the set of vector representations 118A associated with a given cluster from the set of clusters 120.”)(e.g., col 6 lines 56-59 and col 7 lines 10-26) apply a statistical test to the classified data features to determine whether a change between the data features from incoming data is statistically significant compared to historical data features, (“In some implementations, distributions for those additional sets of vector representations can be determined based on the set of clusters 120 and set of statistical properties 122. In some implementations, distributions for those additional sets of vector representations can be determined based on a different set of clusters and statistical properties, such as a set of clusters and associated statistical properties determined based on the set of vector representations 118B and/or a set of vector representations not shown in FIG. 1. As such, the set of clusters and set of statistical properties used for generating the distribution that is to act as the baseline and be compared against another distribution(s) (e.g., from production) for detecting data drift can change over time.”)(e.g., col 9 lines 41-54) the statistical test incorporating a population stability index score; and (“the set of divergence metrics are detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback-Leibler divergence.” – Kullback-Leibler is considered to incorporate a population stability index score)(e.g., col 19 lines 29-32) indicate, based on the population stability index score surpassing a threshold value, that there is a drift in the data features due to the change between the data features from the incoming data being statistically significant compared to the historical data features. (“At 409, data drift is detected between the first set of vector representations and the second set of vector representations based on a comparison of the distribution associated with the first set of vector representations with the distribution associated with the second set of vector representations. In some implementations, the data drift is detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback-Leibler divergence. In some implementations, step 409 is performed automatically (e.g., without requiring additional human input) in response to completing step 408.” “At 410, transmission of a signal is caused, in response to the data drift exceeding a data drift threshold, to cause a remedial action. With reference to FIG. 1, the signal could be transmitted to the data drift detection device 110, user compute device 150, and/or a compute device not shown in FIG. 1.” “At 506, data drift between the first set of vector representations and the second set of vector representations is detected based on a comparison of the distribution associated with the first set of vector representations with the distribution associated with the second set of vector representations at 505. In some implementations, step 506 is performed automatically (e.g., without requiring additional human input) in response to completing step 505.” “At 507, transmission of a signal is caused, in response to the data drift exceeding a data drift threshold, to cause a remedial action. With reference to FIG. 1, the signal could be transmitted to the data drift detection device 110, user compute device 150, and/or a compute device not shown in FIG. 1.”)(e.g., figures 4B and 5 and col 14 lines 20-30 and 31-36 and col 17 lines 10-17 and 18-23) However, Iyer does not appear to specifically disclose classify the data features as either being categorical or numerical; On the other hand, Bareliyahu, which relates to a greedy lookahead K-anonymity for SMB search (title), does disclose classify the data features as either being categorical or numerical; (“For example, numerical data, textual data and categorical data may be masked using different techniques. FIG. 2 shows an example method 200 performed by the disclosed system to provide K-anonymity processing for different data types, based on the principles disclosed herein. As shown in step 202, the method filters the data based on data type. In other words, the type of data in each identifier category is determined by the disclosed system. In one example, categorical data, textual data and numerical data are filtered separately in steps 204A, 204B and 204C for separate K-anonymity processing. In steps 206A, 206B and 206C, the categorical data, textual data and numerical data are respectively processed according to different masking techniques including but not limited to clustering, character subtraction and grouping.” “classifying the entries as either categorical data, textual data or numerical data; and applying the masking function to modify the entries differently depending on the classification.”)(e.g., paragraphs [0023] and claim 5). Iyer discloses data drift detection to improve data integrity for data within a database. E.g., title. Iyer discloses that distributions of data are compared over time to determine data drift, and an alert can be sent to the user as a threshold is exceeded. However, Iyer does not appear to specifically disclose classifying the data features as either being categorical or numerical. On the other hand, Bareliyahu, which relates to a greedy lookahead K-anonymity for SMB search (title), provides that it is known to classify data as either categorical textual or numerical, because it is known that the different data types are processed differently and it is beneficial to identify the data type in order to process the data properly. E.g., paragraph [0023]. This provides a manner to ensure data is properly processed and filtered based on the particular requirements imposed by the particular data tye. E.g., paragraphs [0023]. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of Applicant’s claimed invention to incorporate the determination of the datatype being either numerical or categorical or textual as disclosed in Bareliyahu to Iyer to ensure data is appropriately being processed based on data type by first accurately determining and classifying the data type. Regarding claim 2, Iyer in view of Bareliyahu discloses the computing system of claim 1. Iyer further discloses wherein the executable code, when executed further causes the at least one processor to set the threshold value. (“In some implementations, if dimensionality of the set of vector representations 118A and/or 118B is too high, dimensionality of the set of vector representations 118A and/or 118B can be further reduced until the dimensionality is less than a predetermined threshold (e.g., less than or equal to 5 dimensions, less than or equal to 10 dimensions, less than or equal to 15 dimensions, less than or equal to 20 dimensions, less than or equal to 25 dimensions, etc.). For example, UMAP can be used to perform further dimensionality reduction.” “In some implementations, a vector representation determined to be an outlier can be ignored when generating distributions 124A (and/or distributions 124B). In some implementations, a vector representation is an outlier if a distance between the vector representation and the cluster nearest/most similar to the vector representation in the vector space is greater than a predetermined threshold. In some implementations, a vector representation is an outlier if a distance between the vector representation and the cluster nearest/most similar to the vector representation in the vector space is less than a predetermined threshold.”)(e.g., col 9 lines 22-31 and col 10 lines 46-57). Regarding claim 4, Iyer in view of Bareliyahu discloses the computing system of claim 1. Iyer further discloses wherein the statistical test calculates a respective population stability index score for respective features of the data features. (“the set of divergence metrics are detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback-Leibler divergence.” – Kullback-Leibler is considered to incorporate a population stability index score)(e.g., col 19 lines 29-32). Regarding claim 5, Iyer in view of Bareliyahu discloses the computing system of claim 1. Iyer further discloses wherein the executable code, when executed further causes the at least one processor to determine whether one or more machine learning models rely upon the data features to perform a prediction. (“A distribution 124B associated with the set of data 116B and set of vector representations 118B can be determined based on the set of statistical properties 122. In some implementations, each vector representation from the set of vector representations 118B becomes associated with (e.g., is linked to) a cluster from the set of clusters 120 that is most similar to that vector representation using the set of statistical properties 122. In such a case, the distribution 124B can indicate, for each vector representation from the set of vector representations 118B, a cluster from the set of clusters 120 whose associated statistical property from the set of statistical properties 122 to which that vector representation is most similar. As such, the distribution 124B can also indicate the number of vector representations from the set of vector representations 118B associated with a given cluster from the set of clusters 120. Note that both distributions 124A and 124B were determined based on the same set of statistical properties (i.e., set of statistical properties 122). Additionally, note that one or more clusters from the set of clusters can each be associated with multiple different vector representations from the set of vector representations 118B.”)(e.g., col 7 line 55 – col 8 line 8). Regarding claim 6, Iyer in view of Bareliyahu discloses the computing system of claim 5. Iyer further discloses wherein the executable code, when executed further causes the at least one processor to, based on determining that at least one machine learning model of the one or more machine learning models relies upon the data features, identify one or more administrative users that oversees management of the at least one machine learning model. (“For example, the remedial action could include causing an alert indicating that the data drift exceeds the data drift threshold to be triggered. As another example, the remedial action could include causing a compute device (e.g., data drift detection compute device 110 and/or user compute device 150) and/or machine learning model (e.g., machine learning model 156) to change and/or modify a mode of operation, such as refraining from producing additional output data, refraining from relying on recently produced output data, logging events, shutting down, operating in a restricted access mode, and/or the like. As another example, the remedial action could include causing a machine learning model using the second set of data to be retrained.” – users that are notified are considered to be administrative users that can implement the remedial action.)(e.g., col 14 lines 36-49). Regarding claim 7, Iyer in view of Bareliyahu discloses the computing system of claim 6. Iyer further discloses wherein the executable code, when executed further causes the at least one processor to transmit an electronic notification to respective computing devices associated with the one or more administrative users. (“For example, the data drift detection compute device 110 can send a signal to user compute device 150 and/or a compute device not shown in FIG. 1, alerting the presence of data drift and/or data drift beyond the predetermined data drift threshold.” “For example, the remedial action could include causing an alert indicating that the data drift exceeds the data drift threshold to be triggered.”)(e.g., col 8 lines 25-29 and col 14 lines 36-38). Regarding claim 15, Iyer discloses a computer-implemented method, comprising: (“FIGS. 4A-4B show a flowchart of a method 400 for causing a remedial action in response to detecting a data drift exceeding a data drift threshold, according to an embodiment. In some implementations, method 400 can be performed by a processor (e.g., processor 112).”)(e.g., col 12 lines 59-63) performing data processing on one or more datasets; (“For example, the machine learning model 156 may be used to perform functions related to natural language processing (e.g., analyzing speech, analyzing entities, analyzing sentiment, etc.) or computer vision (e.g., analyzing images, analyzing video, etc.). The machine learning model 156 may be designed and/or trained at any compute device, such as data drift detection compute device 110, user compute device 150, and/or a compute device not shown in FIG. 1.” “the machine learning model 156 is associated with a natural language processing task, and the sets of vector representations 118A and 118B are embedding vectors. Using embedding vectors allows semantic similarity of associated natural langue processing data to be quantified (e.g., by how close the vectors are in vector space).”)(e.g., col 5 lines 28-36 and col 9 lines 55-61) deriving, from the one or more datasets, data features that would be used in data analysis; (“Some techniques described herein generate and use vector representations of higher dimension data and/or unstructured data, such as certain forms of images, images with captions, free form text, multi-modal data, video, audio, social media posts, and/or the like. In some instances, generating vector representations from higher dimension data and/or unstructured data can enable data drift to be detected with respect to that higher dimension data and/or unstructured data, something that may not have been possible if vector representations were not generated.”)(e.g., col 3 lines 65 – col 4 line 7) classifying the data features as either being categorical or numerical ; (“In some implementations, each vector representation from the set of vector representations 118A is associated with (e.g., is linked to, is binned to) a cluster from the set of clusters 120 that is most similar to (e.g., closest distance in vector space) that vector representation. In some instances, multiple different vector representations from the set of vector representations 118A can be associated with a common cluster from the set of clusters 120. In some implementations, the distribution 124A can indicate, for each vector representation from the set of vector representations 118A, a cluster from the set of clusters 120 whose associated statistical property from the set of statistical properties 122 to which that vector representation is most similar. As such, the distribution 124A can also indicate the number of vector representations from the set of vector representations 118A associated with a given cluster from the set of clusters 120.” “In some implementations, each vector representation from the set of vector representations 118A is associated with (e.g., is linked to, is binned to) a cluster from the set of clusters 120 that is most similar to (e.g., closest distance in vector space) that vector representation. In some instances, multiple different vector representations from the set of vector representations 118A can be associated with a common cluster from the set of clusters 120. In some implementations, the distribution 124A can indicate, for each vector representation from the set of vector representations 118A, a cluster from the set of clusters 120 whose associated statistical property from the set of statistical properties 122 to which that vector representation is most similar. As such, the distribution 124A can also indicate the number of vector representations from the set of vector representations 118A associated with a given cluster from the set of clusters 120.”)(e.g., col 6 lines 56-59 and col 7 lines 10-26) applying a statistical test to the classified data features to determine whether a change between the data features from incoming data is statistically significant compared to historical data features, (“In some implementations, distributions for those additional sets of vector representations can be determined based on the set of clusters 120 and set of statistical properties 122. In some implementations, distributions for those additional sets of vector representations can be determined based on a different set of clusters and statistical properties, such as a set of clusters and associated statistical properties determined based on the set of vector representations 118B and/or a set of vector representations not shown in FIG. 1. As such, the set of clusters and set of statistical properties used for generating the distribution that is to act as the baseline and be compared against another distribution(s) (e.g., from production) for detecting data drift can change over time.”)(e.g., col 9 lines 41-54) the statistical test incorporating a population stability index score; and (“the set of divergence metrics are detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback- Leibler divergence.” – Kullback-Leibler is considered to incorporate a population stability index score)(e.g., col 19 lines 29-32) indicating, based on the population stability index score surpassing a threshold value, that there is a drift in the data features due to the change between the data features from the incoming data being statistically significant compared to the historical data features. (“At 409, data drift is detected between the first set of vector representations and the second set of vector representations based on a comparison of the distribution associated with the first set of vector representations with the distribution associated with the second set of vector representations. In some implementations, the data drift is detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback-Leibler divergence. In some implementations, step 409 is performed automatically (e.g., without requiring additional human input) in response to completing step 408.” “At 410, transmission of a signal is caused, in response to the data drift exceeding a data drift threshold, to cause a remedial action. With reference to FIG. 1, the signal could be transmitted to the data drift detection device 110, user compute device 150, and/or a compute device not shown in FIG. 1.” “At 506, data drift between the first set of vector representations and the second set of vector representations is detected based on a comparison of the distribution associated with the first set of vector representations with the distribution associated with the second set of vector representations at 505. In some implementations, step 506 is performed automatically (e.g., without requiring additional human input) in response to completing step 505.” “At 507, transmission of a signal is caused, in response to the data drift exceeding a data drift threshold, to cause a remedial action. With reference to FIG. 1, the signal could be transmitted to the data drift detection device 110, user compute device 150, and/or a compute device not shown in FIG. 1.”)(e.g., figures 4B and 5 and col 14 lines 20-30 and 31-36 and col 17 lines 10-17 and 18-23) However, Iyer does not appear to specifically disclose classifying the data features as either being categorical or numerical; On the other hand, Bareliyahu, which relates to a greedy lookahead K-anonymity for SMB search (title), does disclose classifying the data features as either being categorical or numerical; (“For example, numerical data, textual data and categorical data may be masked using different techniques. FIG. 2 shows an example method 200 performed by the disclosed system to provide K-anonymity processing for different data types, based on the principles disclosed herein. As shown in step 202, the method filters the data based on data type. In other words, the type of data in each identifier category is determined by the disclosed system. In one example, categorical data, textual data and numerical data are filtered separately in steps 204A, 204B and 204C for separate K-anonymity processing. In steps 206A, 206B and 206C, the categorical data, textual data and numerical data are respectively processed according to different masking techniques including but not limited to clustering, character subtraction and grouping.” “classifying the entries as either categorical data, textual data or numerical data; and applying the masking function to modify the entries differently depending on the classification.”)(e.g., paragraphs [0023] and claim 5). It would have been obvious to one of ordinary skill in the art before Applicant’s claimed invention to combine Bareliyahu with Iyer for the reasons provided in claim 1, above. Regarding claim 16, Iyer in view of Bareliyahu discloses the computer-implemented method of claim 15. Iyer further discloses further comprising setting the threshold value. (“In some implementations, if dimensionality of the set of vector representations 118A and/or 118B is too high, dimensionality of the set of vector representations 118A and/or 118B can be further reduced until the dimensionality is less than a predetermined threshold (e.g., less than or equal to 5 dimensions, less than or equal to 10 dimensions, less than or equal to 15 dimensions, less than or equal to 20 dimensions, less than or equal to 25 dimensions, etc.). For example, UMAP can be used to perform further dimensionality reduction.” “In some implementations, a vector representation determined to be an outlier can be ignored when generating distributions 124A (and/or distributions 124B). In some implementations, a vector representation is an outlier if a distance between the vector representation and the cluster nearest/most similar to the vector representation in the vector space is greater than a predetermined threshold. In some implementations, a vector representation is an outlier if a distance between the vector representation and the cluster nearest/most similar to the vector representation in the vector space is less than a predetermined threshold.”)(e.g., col 9 lines 22-31 and col 10 lines 46-57). Regarding claim 18, Iyer in view of Bareliyahu discloses the computer-implemented method of claim 15. Iyer further discloses wherein the statistical test calculates a respective population stability index score for respective features of the data features. (“the set of divergence metrics are detected using at least one of Jensen-Shannon divergence, Earth Mover's Distance, Wasserstein metric, or Kullback-Leibler divergence.” – Kullback-Leibler is considered to incorporate a population stability index score)(e.g., col 19 lines 29-32). Regarding claim 19, Iyer in view of Bareliyahu discloses the computer-implemented method of claim 15. Iyer further discloses further comprising determining whether one or more machine learning models rely upon the data features to perform a prediction. (“A distribution 124B associated with the set of data 116B and set of vector representations 118B can be determined based on the set of statistical properties 122. In some implementations, each vector representation from the set of vector representations 118B becomes associated with (e.g., is linked to) a cluster from the set of clusters 120 that is most similar to that vector representation using the set of statistical properties 122. In such a case, the distribution 124B can indicate, for each vector representation from the set of vector representations 118B, a cluster from the set of clusters 120 whose associated statistical property from the set of statistical properties 122 to which that vector representation is most similar. As such, the distribution 124B can also indicate the number of vector representations from the set of vector representations 118B associated with a given cluster from the set of clusters 120. Note that both distributions 124A and 124B were determined based on the same set of statistical properties (i.e., set of statistical properties 122). Additionally, note that one or more clusters from the set of clusters can each be associated with multiple different vector representations from the set of vector representations 118B.”)(e.g., col 7 line 55 – col 8 line 8). Regarding claim 20, Iyer in view of Bareliyahu discloses the computer-implemented method of claim 16. Iyer further discloses further comprising, based on determining that at least one machine learning model of the one or more machine learning models relies upon the data features, identifying one or more administrative users that oversees management of the at least one machine learning model. (“For example, the remedial action could include causing an alert indicating that the data drift exceeds the data drift threshold to be triggered. As another example, the remedial action could include causing a compute device (e.g., data drift detection compute device 110 and/or user compute device 150) and/or machine learning model (e.g., machine learning model 156) to change and/or modify a mode of operation, such as refraining from producing additional output data, refraining from relying on recently produced output data, logging events, shutting down, operating in a restricted access mode, and/or the like. As another example, the remedial action could include causing a machine learning model using the second set of data to be retrained.” – users that are notified are considered to be administrative users that can implement the remedial action.)(e.g., col 14 lines 36-49) . 07-21-aia AIA Claims 3 and 17 are re jected under 35 U.S.C. 103 as being unpatentable over Iy er in view of Bareliyahu and in further view of Penfield et al. (U.S. Publication No. 2024/0135164 A1, hereinafter referred to as “Penfield”). Re garding claim 3, Iyer in view of Bareliyahu discloses the computing system of claim 1. However, neither reference appears to specifically disclose wherein the executable code, when executed further causes the at least one processor to calculate a ratio of a unique value assigned to a data feature of the data features divided by a non-null value total number of a column of feature data to generate the ratio, wherein a total less than a predefined ratio percentage indicates the data features are to receive a categorical classification, wherein if the total is greater than the predefined ratio percentage the data features are to receive a numerical classification, wherein the classifying is based on calculating the ratio. On the other hand, Penfield, which relates to automatic identification of lessons-learned incident records (title), does disclose wherein the executable code, when executed further causes the at least one processor to calculate a ratio of a unique value assigned to a data feature of the data features divided by a non-null value total number of a column of feature data to generate the ratio, wherein a total less than a predefined ratio percentage indicates the data features are to receive a categorical classification, wherein if the total is greater than the predefined ratio percentage the data features are to receive a numerical classification, wherein the classifying is based on calculating the ratio. (“Another type of identifiable field is the “categorical” field which is a field with a small number of unique textual values. The categorical field may be for example a “consequences” field that categorizes the consequences of an accident with field entries or values “low”, “moderate”, and “high”. A third field type that may be identified by the preprocessing 103 is a “small-vocab” field which is a textual column but one that has too many entries or unique values such as words or phrases to fall within a “categorical” type field, but has a vocabulary (unique words) that is too small to be classified as a “free-text” field. An example would be a field titled “TypeOfEquipmentFailure” which may read “Mechanical failure—gas coolers”. A fourth category or type of field is a “Quantity-based” field that typically contains numeric and textual information describing or capturing area, volume, or time information, generally with the textual entry in this field being limited to describing a unit of measurement. An example of a “Quantity-based” field is a field entry in a document “OnsiteAreaAffected” with a value of “16 sq ft”. A fifth category of field is a “Date and time” field that contains date and time information, an example would include a “Date” field with a value of “2018-06-29 22:00:00”. Other field types or categories are also possible, which have different criteria or parameters, and may be added to the categories disclosed herein, or even replace these, depending on the embodiment.”)(e.g., paragraph [0034]) It would have been obvious to one of ordinary skill in the art before Applicant’s claimed invention to combine Bareliyahu with Iyer for the reasons provided in claim 1, above. Iyer discloses data drift detection to improve data integrity for data within a database. E.g., title. Iyer discloses that distributions of data are compared over time to determine data drift, and an alert can be sent to the user as a threshold is exceeded. However, Iyer does not appear to specifically disclose calculating a unique value ratio to determine whether the data feature is categorical or numerical. On the other hand, Penfield, which relates to automatic identification of lessons-learned incident records (title), provides that it is known that data can be processed and types of fields can be identified by calculating uniqueness of values. E.g., paragraph [0034]. This provides a manner to optimize the manner to identify the type of values that are trained and how to identify changes and optimizes performance of the model. E.g., paragraphs [0037]-[0039]. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of Applicant’s claimed invention to incorporate the determination of numerical and categorical features based on a uniqueness ratio as disclosed in Penfield to the Iyer-Bareliyahu combination to optimize the performance of the system and to improve the manner in which items are identified for subsequent data drift detection analysis. Regarding claim 17, Iyer in view of Bareliyahu discloses the computer-implemented method of claim 15. However, neither reference appears to specifically disclose further comprising calculating a ratio of a unique value assigned to a data feature of the data features divided by a non-null value total number of a column of feature data to generate the ratio, wherein a total less than a predefined ratio percentage indicates the data features are to receive a categorical classification, wherein if the total is greater than the predefined ratio percentage the data features are to receive a numerical classification, wherein the classifying is based on calculating the ratio. On the other hand, Penfield, which relates to automatic identification of lessons-learned incident records (title), does disclose further comprising calculating a ratio of a unique value assigned to a data feature of the data features divided by a non-null value total number of a column of feature data to generate the ratio, wherein a total less than a predefined ratio percentage indicates the data features are to receive a categorical classification, wherein if the total is greater than the predefined ratio percentage the data features are to receive a numerical classification, wherein the classifying is based on calculating the ratio. (“Another type of identifiable field is the “categorical” field which is a field with a small number of unique textual values. The categorical field may be for example a “consequences” field that categorizes the consequences of an accident with field entries or values “low”, “moderate”, and “high”. A third field type that may be identified by the preprocessing 103 is a “small-vocab” field which is a textual column but one that has too many entries or unique values such as words or phrases to fall within a “categorical” type field, but has a vocabulary (unique words) that is too small to be classified as a “free-text” field. An example would be a field titled “TypeOfEquipmentFailure” which may read “Mechanical failure—gas coolers”. A fourth category or type of field is a “Quantity-based” field that typically contains numeric and textual information describing or capturing area, volume, or time information, generally with the textual entry in this field being limited to describing a unit of measurement. An example of a “Quantity-based” field is a field entry in a document “OnsiteAreaAffected” with a value of “16 sq ft”. A fifth category of field is a “Date and time” field that contains date and time information, an example would include a “Date” field with a value of “2018-06-29 22:00:00”. Other field types or categories are also possible, which have different criteria or parameters, and may be added to the categories disclosed herein, or even replace these, depending on the embodiment.”)(e.g., paragraph [0034]) It would have been obvious to one of ordinary skill in the art before Applicant’s claimed invention to combine Penfield, Bareliyahu and Iyer for the reasons provided in claim 3, above. Conclusion 07-96 The prior art made of record, listed on form PTO-892, and not relied upon is considered pertinent to applicant's disclosure. Any inquiry concerning this communication or earlier communications from the examiner should be directed to RICHARD L BOWEN whose telephone number is (571)270-5982. The examiner can normally be reached Monday through Friday 7:30AM - 4:00PM 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, Aleksandr Kerzhner can be reached at (571)270-1760. 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. /RICHARD L BOWEN/ Primary Examiner, Art Unit 2165 Application/Control Number: 18/401,813 Page 2 Art Unit: 2165 Application/Control Number: 18/401,813 Page 3 Art Unit: 2165 Application/Control Number: 18/401,813 Page 4 Art Unit: 2165 Application/Control Number: 18/401,813 Page 5 Art Unit: 2165 Application/Control Number: 18/401,813 Page 6 Art Unit: 2165 Application/Control Number: 18/401,813 Page 7 Art Unit: 2165 Application/Control Number: 18/401,813 Page 8 Art Unit: 2165 Application/Control Number: 18/401,813 Page 9 Art Unit: 2165 Application/Control Number: 18/401,813 Page 10 Art Unit: 2165 Application/Control Number: 18/401,813 Page 11 Art Unit: 2165 Application/Control Number: 18/401,813 Page 12 Art Unit: 2165 Application/Control Number: 18/401,813 Page 13 Art Unit: 2165 Application/Control Number: 18/401,813 Page 14 Art Unit: 2165 Application/Control Number: 18/401,813 Page 15 Art Unit: 2165 Application/Control Number: 18/401,813 Page 16 Art Unit: 2165 Application/Control Number: 18/401,813 Page 17 Art Unit: 2165 Application/Control Number: 18/401,813 Page 18 Art Unit: 2165 Application/Control Number: 18/401,813 Page 19 Art Unit: 2165 Application/Control Number: 18/401,813 Page 20 Art Unit: 2165 Application/Control Number: 18/401,813 Page 21 Art Unit: 2165 Application/Control Number: 18/401,813 Page 22 Art Unit: 2165 Application/Control Number: 18/401,813 Page 23 Art Unit: 2165 Application/Control Number: 18/401,813 Page 24 Art Unit: 2165 Application/Control Number: 18/401,813 Page 25 Art Unit: 2165 Application/Control Number: 18/401,813 Page 26 Art Unit: 2165 Application/Control Number: 18/401,813 Page 27 Art Unit: 2165 Application/Control Number: 18/401,813 Page 28 Art Unit: 2165 Application/Control Number: 18/401,813 Page 29 Art Unit: 2165 Application/Control Number: 18/401,813 Page 30 Art Unit: 2165 Application/Control Number: 18/401,813 Page 31 Art Unit: 2165 Application/Control Number: 18/401,813 Page 32 Art Unit: 2165
Prosecution Insights