DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 02/26/2026 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Response to Amendment
The amendments filed 03/16/2026 have been entered.
Claims 1-2, 5-9, 12-16 and 19-21 remain pending within the application.
The amendments filed 03/16/2026 are sufficient to overcome each and every objection previously set forth in the Non-Final Office Action mailed 12/15/2025. The objections have been withdrawn.
The amendments filed 03/16/2026 are sufficient to overcome the 112(b) rejections previously set forth in the Non-Final Office Action mailed 12/15/2025. The rejections have been withdrawn.
The amendments and remarks filed 03/16/2026 are sufficient to overcome the 101 rejections previously set forth in the Non-Final Office Action mailed 12/15/2025. The rejections have been withdrawn. Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action.
The amendments and remarks filed 03/16/2026 are sufficient to overcome the 102/103 rejections previously set forth in the Non-Final Office Action mailed 12/15/2025. The rejections have been withdrawn. Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action.
Claim Objections
Claim 1 is objected to because of the following informalities: “a number hard negative types” should be: “a number of hard negative types”. Appropriate correction is required.
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.
The claims 1, 2, 5-9, 12-16, and 19-21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 2A Prong 1: Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Claim 1 includes the steps of:
An apparatus comprising: machine-readable instructions; and at least one programmable circuit to instantiate: block circuitry to:
generate a plurality of blockings based on first data samples, the plurality of blockings including a first blocking corresponding to first ones of the first data samples
match circuitry to: identify the first blocking as a similar to a second data sample based on the second data sample being associated with the first heuristic, the second data sample associated with a plurality of heuristics, the plurality of heuristics including the first heuristic; and
compare values of the plurality of heuristics of the second data sample to respective values of a set of heuristics associated with the first blocking to
identify (a) the first ones of the first data samples for which the second data sample is a positive match type and (b) the first ones of the first data samples for which the second data sample is a hard negative type, the hard negative type having a degree of similarity different than the positive match type;
threshold evaluation circuitry to: determine if a number hard negative types satisfies a threshold amount,
the match circuitry to, responsive to the number of hard negative types failing to satisfy the threshold amount, discard the second data sample and the first blocking; and
batch circuitry to: responsive to the number of hard negative types satisfying the threshold amount, combine the first ones of the first data samples corresponding to the positive match type and the first ones of the first data samples corresponding to the hard negative type into a machine learning input batch; and
cause a machine learning model to be trained based on the machine learning input batch.
The broadest reasonable interpretation of the bolded limitations above are directed to a mental process able to be performed in the human mind through the use of a physical aid, like a pen and paper. A human can:
generate blockings of data from data samples associated with various heuristics,
identify a match of blockings based on whether the data sample includes a heuristic,
compare heuristics to data samples,
identify data samples which are a positive match type and hard negative type,
determine if hard negative types fail to satisfy a threshold amount and discard data samples and blockings as a result,
determine if hard negative types satisfy a threshold amount and combine data samples into an input batch as a result.
Step 2A Prong 2: Does the claim recite additional elements that integrate the judicial exception into a practical application?
As drafted and under their broadest reasonable interpretation, the following limitations recite additional elements which amount to generic computer components recited at a high level of generality, with merely the words “apply it” or an equivalent with the judicial exception, merely including instructions to implement an abstract idea on the additional elements, or merely using the additional elements as a tool to perform an abstract idea, as discussed in MPEP 2106.05(f).
An apparatus comprising: machine-readable instructions; and at least one programmable circuit to instantiate…
block circuitry,
match circuitry,
batch circuitry,
threshold evaluation circuitry,
cause a machine learning model to be trained based on the machine learning input batch.
As drafted and under their broadest reasonable interpretation, the following limitations recite additional elements which amount to mere data gathering and output recited at a high level of generality, and thus are insignificant extra-solution activity. See MPEP 2106.05(g) (“whether the limitation is significant”). In addition, all uses of the recited judicial exceptions require such data gathering and output, and, as such, these limitations do not impose any meaningful limits on the claim. These limitations amount to necessary data gathering and outputting. See MPEP 2106.05.
cause a machine learning model to be trained based on the machine learning input batch.
The additional elements have been considered both individually and as an ordered combination in order to determine whether they integrates the exception into a practical application. Therefore, no meaningful claim limits are imposed practicing the abstract idea. Accordingly, at Step 2A, prong two, the additional elements do not integrate the judicial exception into a practical application.
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
The claim limitation(s) reciting generic computer elements amounts to no more than mere instructions to apply the exception using a generic computer.
The claim reciting the additional element(s) of machine learning input amount to necessary data gathering and output.
The additional elements have been considered both individually and as an ordered combination in order to determine whether they warrant significantly more consideration. Thus, the claim does not provide an inventive concept.
The claim is ineligible.
Claims 2 and 5-7 further recite limitations that encompass mental evaluations that are practically performed in the human mind, but for the recitation of generic computer components, and additional element(s) of “retrieving” amounting to necessary data gathering and output. The claims do not integrate the judicial exception into practical application. The claims do not recite additional elements that are sufficient to amount to significantly more than the judicial exception.
Claims 2 and 5-7 are ineligible.
Claims 8, 9, 12-14 and 15, 16, 19-21 are substantially similar to claims 1, 2, 5-7 respectively, and are rejected on the same basis as claims 1, 2, 5-7. These claims further recite additional elements that amount to generic computer components recited at a high level of generality, with merely the words “apply it” or an equivalent with the judicial exception, merely including instructions to implement an abstract idea on the additional elements, or merely using the additional elements as a tool to perform an abstract idea.
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, 5, 8, 12, 15, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Erenrich et al. (Pub. No.: US 2019/0188200 A1), hereafter Erenrich, in view of Cheng et al. (Pub. No.: US 2025/0005460 A1), prior art made record of and not relied on in the office action mailed 12/15/2025, hereafter Cheng.
Regarding claim 1, Erenrich discloses:
An apparatus comprising: machine-readable instructions; and at least one programmable circuit to instantiate: (Fig. 4, ¶[0043-0056]):
block circuitry to: generate a plurality of blockings based on first data samples, the plurality of blockings including a first blocking corresponding to first ones of the first data samplesassociated with a first heuristic (Fig. 5 element 520, Fig. 6 element 620, ¶[0062] and ¶[0092] teaches generating a plurality of blockings, i.e. one or more grouping of records, including a first blocking corresponding to a first subset of data samples associated with fields, i.e. a first heuristic),
match circuitry to: identify the first blocking as a similar to a second data sample based on the second data sample being associated with the first heuristic, the second data sample associated with a plurality of heuristics, the plurality of heuristics including the first heuristic (Fig. 6 element 640, ¶[0068], Fig. 2, and ¶[0063] teaches identifying the first blocking, i.e. group, as similar to a second data sample in the second group based on the second data sample being associated with the first heuristic, i.e. field, and the second data sample being associated with one or more fields that include the field of the first group),
compare values of the plurality of heuristics of the second data sample to respective values of a set of heuristics associated with the first blocking to identify (a) the first ones of the first data samples for which the second data sample is a positive match type and (b) the first ones of the first data samples for which the second data sample is a hard negative type, the hard negative type having a degree of similarity different than the positive match type (Fig. 5, ¶[0079], and ¶[0072] teaches comparing the heuristics of the second data sample to the heuristics of the first blocking to identify first data samples for which the second data sample is a positive match type, i.e. matching, and identifying first data samples for which the second data sample is a hard negative type, i.e. non-matching, where matching and non-matching implies a different degree of similarity),
threshold evaluation circuitry to: determine if a number hard negative types satisfies a threshold amount (¶[0087] teaches determining if the hard negative types in pairs satisfy a certain threshold during assessment),
the match circuitry to, responsive to the number of hard negative types failing to satisfy the threshold amount, discard the second data sample … (¶[0087] teaches filtering to discard samples that fail to satisfy a certain threshold).
batch circuitry to: responsive to the number of hard negative types satisfying the threshold amount, combine the first ones of the first data samples corresponding to the positive match type and the first ones of the first data samples corresponding to the hard negative type into a machine learning input batch (¶[0080] and ¶[0090] teaches inputting the combined matching and non-matching samples to the machine learning model, where global optimization combines these samples after filtering, responsive to the number of hard negative types satisfying the threshold amount)
cause a … learning model to be trained based on the machine learning input batch (Fig. 5 and ¶[0076] teaches training a statistical learning model based on a machine learning model and its input batch).
While Erenrich discloses responsive to the number of hard negative types failing to satisfy the threshold amount, discard the second data sample, and cause a … learning model to be trained based on the machine learning input batch, they do not explicitly disclose discarding the first blocking responsive to failing to satisfy a threshold amount and training a machine learning model.
Cheng discloses:
responsive to … failing to satisfy the threshold amount, discard … the first blocking (Fig. 2, elements “S_3.5” and “selected blocking model”, and ¶[0053] teaches discarding blockings that fail to meets a threshold F1 score),
cause a machine learning model to be trained based on the machine learning input batch (Fig. 1 and ¶[0037] teaches training and retraining the machine learning models based on a machine learning input batch).
Erenrich and Cheng are analogous art because they are from the same field of endeavor, entity matching and machine learning models.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Erenrich to include responsive to … failing to satisfy the threshold amount, discard … the first blocking and cause a machine learning model to be trained based on the machine learning input batch, based on the teachings of Cheng. One of ordinary skill in the art would have been motivated to make this modification in order to reduce the time and cost required for entity matching, as suggested by Cheng (¶[0019]).
Regarding claim 5, Erenrich, in view of Cheng, discloses the apparatus as defined in claim 1 (and thus the rejection of claim 1 is incorporated). Erenrich further discloses:
wherein the block circuitry is to retrieve the first data samples from a first data source and the second data sample from a second data source, the second data source including the first data source (Fig. 7, elements 710, ¶[0018], and ¶[0063] teaches retrieving the first data samples from a first data source. i.e. first list, and the second data sample from a second data source, i.e. second list, where the first and second list are to be represented from the same list in the data storage device, and thus the second data source includes the first data source).
Claims 8 and 15 are substantially similar to claim 1 , and are rejected on the same basis as claim 1.
Claims 12 and 19 are substantially similar to claim 5, and are rejected on the same basis as claim 5.
Claims 2, 6-7, 9, 13-14, 16, and 20-21 are rejected under 35 U.S.C. 103 as being unpatentable over Erenrich et al. (Pub. No.: US 2019/0188200 A1), hereafter Erenrich, in view of Cheng et al. (Pub. No.: US 2025/0005460 A1), prior art made record of and not relied on in the office action mailed 12/15/2025, hereafter Cheng, in further view of Sakai et al. ("Entity Matching with String Transformation and Similarity-Based Features"), hereafter Sakai.
Regarding claim 2, Erenrich, in view of Cheng, discloses the apparatus as defined in claim 1 (and thus the rejection of claim 1 is incorporated). Erenrich further discloses:
wherein the plurality of blockings includes a second blocking corresponding to second ones of the first data samples that are associated with a second heuristic (Fig. 5, element 525 and ¶[0066] teaches a second blocking corresponding to second ones of the first data samples that are associated with a second heuristic, i.e. field),
the match circuitry is to: compare the first blocking against the second blocking; and assign respective ones of the first data samples a … negative type (Fig. 5, ¶[0079], and ¶[0072] teaches comparing pairs to assign non-matching, negative type labels),
the batch circuitry to: add the ones of the first data samples corresponding to the … negative type into the machine learning input batch (¶[0080] teaches adding the non-matching label samples to the machine learning model input batch).
While Erenrich discloses the match circuitry is to: compare the first blocking against the second blocking; and assign respective ones of the first data samples a … negative type, and
the batch circuitry to: add the ones of the first data samples corresponding to the … negative type into the machine learning input batch, they do not explicitly disclose this negative type to be an easy negative type.
Sakai discloses:
assign … data samples an easy negative type (page 80, first 4 lines and page 82, first 2 lines teach assigning easy negative types to data samples),
add the ones of the first data samples corresponding to the easy negative type into the machine learning input batch (page 81 last paragraph to page 82 first paragraph “we propose to select the training data by weighted sampling with four well-tuned similarity thresholds. … Easy-negative samples…” teaches adding easy negative type samples to machine learning input batches).
Erenrich, Cheng, and Sakai are analogous art because they are from the same field of endeavor, entity matching and machine learning models.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Erenrich, in view of Cheng, to include assign … data samples an easy negative type, and add the ones of the first data samples corresponding to the easy negative type into the machine learning input batch, based on the teachings of Sakai. One of ordinary skill in the art would have been motivated to make this modification in order to deal with the problem of the hard-to-classify pairs, as suggested by Sakai (page 77, last 2 lines).
Regarding claim 6, Erenrich, in view of Cheng, in further view of Sakai, discloses the apparatus as defined in claim 2 (and thus the rejection of claim 2 is incorporated). Erenrich further discloses:
wherein the first ones of the first data samples and the second ones of the first data samples are labeled with the first heuristic and the second heuristic, respectively (¶[0016] teaches the first and second list to be labeled with distinct ID fields, i.e. respective heuristics).
Regarding claim 7, Erenrich, in view of Cheng, in further view of Sakai, discloses the apparatus as defined in claim 6 (and thus the rejection of claim 6 is incorporated). Erenrich further discloses:
wherein the first heuristic or the second heuristic includes one of brand, product identifier, color, price, small price difference, date sold, or retailer (Fig. 3 teaches consumer heuristics such as price and date sold).
Claims 9 and 16 are substantially similar to claim 2, and are rejected on the same basis as claim 2.
Claims 13-14 and 20-21 are substantially similar to claims 6-7, and are rejected on the same basis as claims 6-7.
Response to Arguments
Applicant's arguments filed 03/16/2026 have been fully considered with regards to the 35 U.S.C. 101 rejection, but they are not persuasive.
The applicant asserts on page 15-16 of the remarks “the claimed subject matter of the instant application is directed to a specific manner for generating training data to train a machine learning model…claim 1 of the present application also expressly reflects the improvement of including both positive matches and hard negatives in training data…Claim 1 has no mention whatsoever of making evaluations and judgements of observations for formulating observations, evaluations, and judgements and, thus, does not set forth a mental process.”. The Examiner respectfully disagrees, as the steps for generating training data are clearly directed to abstract ideas, such as making comparisons, identifications, and determinations (see 101 rejection above). The improvement to the technology must result from the extra-solution activity when an abstract idea is present in the claim, as Step 2A, Prong 2 looks to the extra-solution activity to determine if there is an inventive concept present. For example, in Trading Technologies Int' l v. IBG, 921 F.3d 1084, 1093-94, 2019 USPQ2d 138290 (Fed. Cir. 2019), the court determined that the claimed user interface simply provided a trader with more information to facilitate market trades, which improved the business process of market trading but did not improve computers or technology (MPEP 2106.05(a)(II)). Thus, even when considering the elements in combination, the claim as a whole does not integrate the recited exception into a practical application.
Furthermore, In reference to the applicant’s arguments regarding Desjardins, the applicant is reminded that the claims do not recite an improvement to training a machine learning model or adjustment of model training parameters, as discussed in Desjardins. More specifically, “xiii. An improved way of training a machine learning model that protected the model’s knowledge about previous tasks while allowing it to effectively learn new tasks; Ex Parte Desjardins, Appeal No. 2024-000567 (PTAB September 26, 2025, Appeals Review Panel Decision) (precedential); and xiv. Improvements to computer component or system performance based upon adjustments to parameters of a machine learning model associated with tasks or workstreams; Ex Parte Desjardins, Appeal No. 2024-000567 (PTAB September 26, 2025, Appeals Review Panel Decision) (precedential).” The applicant’s claim are not directed to an improvement in the operation or training of a machine learning model, but rather, various data matching and combination steps that are performable by the human mind using pen and paper.
In reference to the applicant’s arguments comparing examples from the Patent Eligibility Guidance, these examples are hypothetical and only intended to be illustrative of a claim analysis performed using MPEP 2106. The 2024 Patent Eligibility Guidance states “These examples should be interpreted based on the fact patterns set forth below, as other fact patterns may have different eligibility outcomes. That is, it is not necessary for a claim under examination to mirror an example claim to be subject matter eligible. All claims are analyzed for eligibility in accordance with their broadest reasonable interpretation.”. The examiner maintains that the applicants claims have been analyzed for eligibility in accordance to their broadest reasonable interpretation. The claims recite combining data samples, which can be performed by mental evaluation, with the aid of pen and paper. A human could reasonably combine two data samples, for example, two different labels, to consider them to be a singular label.
Applicant's arguments filed 03/16/2026 have been fully considered with regards to the 35 U.S.C. 102/103 rejection.
The applicant argues, in pages 18-19 of the remarks, that Erenrich does not disclose “match circuitry to, responsive to a number of hard negative types failing to satisfy a threshold amount, discard a second data sample and a first blocking; and batch circuitry to responsive to the number of hard negative types satisfying the threshold amount, combine first ones of first data samples corresponding to the positive match type and the first ones of the first data samples corresponding to the hard negative type into a machine learning input batch. None of the cited references teaches or suggests such circuitry.”. However, Erenrich at least discloses - the match circuitry to, responsive to the number of hard negative types failing to satisfy the threshold amount, discard the second data sample … (¶[0087] teaches filtering to discard samples that fail to satisfy a certain threshold), and batch circuitry to: responsive to the number of hard negative types satisfying the threshold amount, combine the first ones of the first data samples corresponding to the positive match type and the first ones of the first data samples corresponding to the hard negative type into a machine learning input batch (¶[0080] and ¶[0090] teaches inputting the combined matching and non-matching samples to the machine learning model, where global optimization combines these samples after filtering, responsive to the number of hard negative types satisfying the threshold amount), but fails to teach discarding the first blocking responsive to failing to satisfy a threshold amount, which is taught by the new reference Cheng as demonstrated in the new grounds of rejection over Erenrich and Cheng. See 103 rejection above.
Applicant’s further arguments with respect to claim(s) 1, 2, 5-9, 12-16, and 19-21 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HUMAIRA ZAHIN MAUNI whose telephone number is (703)756-5654. The examiner can normally be reached Monday - Friday, 9 am - 5 pm (ET).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, MATT ELL can be reached at (571) 270-3264. 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.
/H.Z.M./Examiner, Art Unit 2141
/ANDREW L TANK/Primary Examiner, Art Unit 2141
Prosecution Insights