Prosecution Insights
Last updated: April 19, 2026
Application No. 18/148,058

SYSTEMS AND METHODS FOR USING MACHINE LEARNING TECHNIQUES TO PREDICT ITEM GROUP COMPOSITION

Final Rejection §102
Filed
Dec 29, 2022
Examiner
SYED, FARHAN M
Art Unit
2161
Tech Center
2100 — Computer Architecture & Software
Assignee
Fidelity Information Services LLC
OA Round
2 (Final)
75%
Grant Probability
Favorable
3-4
OA Rounds
3y 9m
To Grant
98%
With Interview

Examiner Intelligence

Grants 75% — above average
75%
Career Allow Rate
621 granted / 829 resolved
+19.9% vs TC avg
Strong +23% interview lift
Without
With
+23.4%
Interview Lift
resolved cases with interview
Typical timeline
3y 9m
Avg Prosecution
29 currently pending
Career history
858
Total Applications
across all art units

Statute-Specific Performance

§101
16.8%
-23.2% vs TC avg
§103
46.1%
+6.1% vs TC avg
§102
20.8%
-19.2% vs TC avg
§112
4.8%
-35.2% vs TC avg
Black line = Tech Center average estimate • Based on career data from 829 resolved cases

Office Action

§102
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 . Status of Claims In response to communications filed on 09 January 2026, claims 1-20 are presently pending in the application, of which, claims 1, 16 and 20 are presented in independent form. The Examiner acknowledges amended claims 17-19. Claims 21-40 were previously cancelled. No claims were newly added. Response to Remarks/Arguments All objections and/or rejections issued in the previous Office Action, mailed 11 September 2025, have been withdrawn, unless otherwise noted in this Office Action. Applicant's arguments filed 09 January 2026 have been fully considered but they are not persuasive. Applicant argues: Brosamer does not teach or suggest the feature of ‘receiving entity identification information and a timestamp associated with a transaction without receiving information distinguishing items associated with the transactions,’ as recited in independent claim 1 and similarly in independent claim 16. The Examiner disagrees. Brosamer teaches receiving entity identification information (e.g. Brosamer, see further column 9, lines 40-55, which discloses POS devices receives transaction information and a card network associated with the payment instrument, which indicates entity identification) and a timestamp associated with a transaction (e.g. Brosamer, see column 9, lines 40-55, which discloses a time, a place, and a date of the transaction, which the Examiner notes are characteristics of a timestamp.) without receiving information distinguishing items associated with the transactions (e.g. Brosamer, see column 9, lines 40-55, which discloses an offline transaction, where the POS devices may store one or more characteristics associated with the transaction, however the transaction does not indicate the items acquired with the transaction that the Examiner interprets as distinguishing items associated with the transaction.). As such, it appears to the Examiner that Brosamer discloses the claimed feature. Brosamer does not teach or suggest the feature of ‘applying the localized machine learning model to a model input to generate predicted categories of items associated with the transaction, the model input including the received entity identification information and a timestamp but not including information distinguishing items associated with the transaction,’ as recited in independent claim 1 and similarly in independent claim 16. The Examiner disagrees. Brosamer teaches applying the localized machine learning model to a model input to generate predicted categories of items associated with the transaction (Brosamer, see column 3, line 32 to column 4, line 60, which discloses a POS device of the merchant as a local machine where merchant category code (MCC) and jobs to be done (JTBD) clusters are collected by the merchant, where the payment processing system can generate training data comprising (a) stored historical transaction data; (b) an indication, for each historical purchase transaction, and (c) a classification of a merchant associated with the transaction and/or business.), the model input including the received entity identification information and a timestamp but not including information distinguishing items associated with the transaction (e.g. Brosamer, see column 2, lines 36-40, which discloses classify merchants into machine classifications by comparing information associated with the merchant using various machine-learning models which may not completely rely on any rules-based programming instructions or algorithms. Additionally, see column 3, line 32 to column 4, line 20, which discloses machine learning models are used to create cluster sellers, in which the learning models can be used to influence the context by choosing what classes are learned to select and predict merchant type and take the learned vectors for more general use.). As such, it appears to the Examiner that Brosamer discloses the claimed feature. Brosamer does not teach or suggest the features of claim 6. The Examiner disagrees. Brosamer teaches aggregating receipt-level transaction data from multiple entities (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which machine learning powered payment service, similar to payment service starts with raw data and/or system level pre-aggregated data to perform additional data transformation using user-defined functions.); imputing values to a portion of the receipt-level transaction data (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which allows for a random forest classification model that predicts ‘food business category’ based on real-time snapshots and aggregated data features.); removing outlier values from the receipt-level transaction data (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which discloses reclassification of sellers can be accomplished by a random forest classification model that predicts food business category based on real-time snapshots and aggregated data features obtained from payment activity.); or applying price constraints or basket constraints to receipt-level transaction data (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which discloses reclassification of sellers can be accomplished by a random forest classification model that predicts food business category based on real-time snapshots and aggregated data features obtained from payment activity.). Brosamer does not teach or suggest the features of claim 11. The Examiner disagrees. Brosamer teaches wherein applying the localized machine learning model to the model input generates at least one predicted brand associated with the transaction (e.g. (Brosamer, see column 3, line 32 to column 4, line 60, which discloses a POS device of the merchant as a local machine where merchant category code (MCC) and jobs to be done (JTBD) clusters are collected by the merchant, where the payment processing system can generate training data comprising (a) stored historical transaction data; (b) an indication, for each historical purchase transaction, and (c) a classification of a merchant associated with the transaction and/or business.). See further Brosamer, see column 24, lines 43-56 and Figure 10, which discloses reclassification of sellers can be accomplished by a random forest classification model that predicts food business category based on real-time snapshots and aggregated data features obtained from payment activity.). Brosamer does not teach or suggest the feature of ‘initializing an untrained machine learning model,’ as recited in independent claim 20. The Examiner disagrees. Brosamer teaches initializing an untrained machine learning model (Brosamer, see column 3, line 32 to column 4, line 60, which discloses a POS device of the merchant as a local machine where merchant category code (MCC) and jobs to be done (JTBD) clusters are collected by the merchant, where the payment processing system can generate training data comprising (a) stored historical transaction data; (b) an indication, for each historical purchase transaction, and (c) a classification of a merchant associated with the transaction and/or business. The Examiner notes that the machine learning model at any point in time can be untrained until trained data is generated from historical transaction data.). Brosamer does not teach or suggest the feature of ‘validating the trained machine learning model by inputting model validation data to the trained machine learning model, the model validation data comprising second receipt-level transaction data not included in the first receipt-level transaction data,’ as recited in independent claim 20. The Examiner disagrees. Brosamer teaches validating the trained machine learning model by inputting model validation data to the trained machine learning model (Brosamer, see column 3, line 32 to column 4, line 60, which discloses a POS device of the merchant as a local machine where merchant category code (MCC) and jobs to be done (JTBD) clusters are collected by the merchant, where the payment processing system can generate training data comprising (a) stored historical transaction data; (b) an indication, for each historical purchase transaction, and (c) a classification of a merchant associated with the transaction and/or business.), the model validation data comprising second receipt-level transaction data not included in the first receipt-level transaction data (e.g. Brosamer, see column 11, lines 28-39 and Figure 1, which discloses generating each of the class profiles, the payment processing system uses the data that represents commonly known information about the classes and/or the data that the payment processing system receives from the merchants. For instance the payment processing system can generate a class profile for grocery stores and supermarkets using data that represents commonly known information about grocery stores and supermarkets.). No other argument was presented by the Applicant and therefore the Examiner maintains the rejection below. Allowable Subject Matter Claims 7 and 19 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. Claim Rejections - 35 USC § 102 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 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 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 – (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. (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. Claims 1-6, 8-18, and 20 are rejected under 35 U.S.C. 102(a)(1)/(a)(2) as being unpatentable by Brosamer, Jacquelyn et al (U.S. 10,949,825 and known hereinafter as Brosamer). As per claim 1, Brosamer teaches a system for predicting item group composition, the system comprising: at least one processor (e.g. Brosamer, see Figure 1, which discloses a processor coupled to memory.); and a non-transitory computer-readable medium containing instructions that, when executed by the at least one processor (e.g. Brosamer, see Figure 1, which discloses a processor coupled to memory.), cause the at least one processor to perform operations comprising: receiving entity identification information and a timestamp associated with a transaction without receiving information distinguishing items associated with the transaction (e.g. Brosamer, see column 20, line 59 to column 21, line 3 and Figure 7, which discloses the heat map of all merchant payment transactions by weekday and time illustrates how local timestamps of transactions can be used as a feature.); determining, based on the entity identification information, a localized machine learning model trained to predict categories of items based on transaction information applying to all of the items associated with the transaction (e.g. Brosamer, see column 2, lines 36-40, which discloses classify merchants into machine classifications by comparing information associated with the merchant using various machine-learning models which may not completely rely on any rules-based programming instructions or algorithms); and applying the localized machine learning model to a model input to generate predicted categories of items associated with the transaction, the model input including the received entity identification information and a timestamp but not including information distinguishing items associated with the transaction (e.g. Brosamer, see column 24, lines 43-56, which discloses model has an average accuracy of 65% across business categories, where categories with very consistent seller activity are easy to predict while more nebulous categories are harder to predict.). As per claim 16, Brosamer teaches a method for predicting item group composition, comprising: receiving entity identification information and a timestamp associated with a transaction without receiving information distinguishing items associated with the transaction (e.g. Brosamer, see column 20, line 59 to column 21, line 3 and Figure 7, which discloses the heat map of all merchant payment transactions by weekday and time illustrates how local timestamps of transactions can be used as a feature.); determining, based on the entity identification information, a localized machine learning model trained to predict categories of items based on transaction information applying to all of the items associated with the transaction (e.g. Brosamer, see column 2, lines 36-40, which discloses classify merchants into machine classifications by comparing information associated with the merchant using various machine-learning models which may not completely rely on any rules-based programming instructions or algorithms); and applying the localized machine learning model to a model input to generate predicted categories of items associated with the transaction, the model input including the received entity identification information and a timestamp but not including information distinguishing items associated with the transaction (e.g. Brosamer, see column 24, lines 43-56, which discloses model has an average accuracy of 65% across business categories, where categories with very consistent seller activity are easy to predict while more nebulous categories are harder to predict.). As per claim 20, Brosamer teaches a system for training an item group composition prediction model, the system comprising: at least one processor (e.g. Brosamer, see Figure 1, which discloses a processor coupled to memory.); and a non-transitory computer-readable medium containing instructions that, when executed by the at least one processor (e.g. Brosamer, see Figure 1, which discloses a processor coupled to memory.), cause the at least one processor to perform operations comprising: initializing an untrained machine learning model (e.g. Brosamer, column 4, line 35 to column 5, line 15, which discloses the payment processing system can generate training data comprising (a) stored historical transaction data, (b) an indication, for each historical purchase transaction, of whether the purchase transaction was ultimately determined to be fraudulent, and (c) a classification of merchant associated with the transaction and/or business, where the collected data can indicate transactional information.); training the untrained machine learning model to predict item categories by: inputting model training data to the untrained machine learning model, the model training data comprising first receipt-level transaction data and categories of items associated with the first receipt-level transaction data, the model training data being received from multiple distinct entities (e.g. Brosamer, see column 11, lines 28-39 and Figure 1, which discloses generating each of the class profiles, the payment processing system uses the data that represents commonly known information about the classes and/or the data that the payment processing system receives from the merchants. For instance the payment processing system can generate a class profile for grocery stores and supermarkets using data that represents commonly known information about grocery stores and supermarkets.); and modifying at least one parameter of the untrained machine learning model to improve prediction of the categories of the items (e.g. Brosamer, see column 18, lines 5-67, which discloses classification component may utilize machine learning mechanism to train the data model, where the data model may be trained using supervised learning algorithm and stored in cluster.); and validating the trained machine learning model by inputting model validation data to the trained machine learning model, the model validation data comprising second receipt-level transaction data not included in the first receipt-level transaction data (e.g. Brosamer, see column 11, lines 28-39 and Figure 1, which discloses generating each of the class profiles, the payment processing system uses the data that represents commonly known information about the classes and/or the data that the payment processing system receives from the merchants. For instance the payment processing system can generate a class profile for grocery stores and supermarkets using data that represents commonly known information about grocery stores and supermarkets.). As per claim 2, Brosamer teaches the system of claim 1, wherein the determined localized machine learning model comprises at least one of an Extreme Gradient Boosting (XG-Boost) model, a random forest model, or a deep learning model (e.g. Brosamer, see column 4, lines 5-21, which discloses deep learning models can be used to influence the context by choosing what classes are learned to select, for example by using a soft max versus metric learning approach.). As per claim 3, Brosamer teaches the system of claim 1, wherein: the operations further comprise receiving a total transaction amount associated with the transaction, without receiving information distinguishing items associated with the transaction (e.g. Brosamer, see column 9, lines 22-40, which discloses when paying for a transaction, the customer can provide the amount that is due to the merchant using a payment instrument.); and the model input further includes the total transaction amount but does not include information distinguishing items associated with the transaction (e.g. Brosamer, see column 9, lines 22-40, which discloses when paying for a transaction, the customer can provide the amount that is due to the merchant using a payment instrument, such as a debit card, a credit card, a stored-value, or gift card, a check through which an electronic payment application on a device carried by the customer.). As per claims 4 and 17, Brosamer teaches the system of claim 1 and the method of claim 16, respectively, wherein: the entity identification information identifies a first store (e.g. Brosamer, see column 18, lines 5-67, which discloses classification component may utilize machine learning mechanism to train the data model, where the data model may be trained using supervised learning algorithm and stored in cluster.); and the localized machine learning model is trained to predict categories of items based on data received from at least one second store, the at least one second store being in a same cluster as the first store (e.g. Brosamer, see column 11, lines 28-39 and Figure 1, which discloses generating each of the class profiles, the payment processing system uses the data that represents commonly known information about the classes and/or the data that the payment processing system receives from the merchants. For instance the payment processing system can generate a class profile for grocery stores and supermarkets using data that represents commonly known information about grocery stores and supermarkets.). As per claims 5 and 18, Brosamer teaches the system of claim 4 and the method of claim 17, respectively, wherein the first store and the at least one second store are clustered in a same cluster based on application of a clustering model to first attribute data received from the first store and second attribute data received from the at least one second store (e.g. Brosamer, see column 20, lines 38-40 and Figure 6B, which discloses an unsupervised clustering model using the contextual features as inputs to generate the most probable cluster as output.). As per claim 6, Brosamer teaches the system of claim 5, wherein the clustering model is generated by performing at least one of: aggregating receipt-level transaction data from multiple entities (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which machine learning powered payment service, similar to payment service starts with raw data and/or system level pre-aggregated data to perform additional data transformation using user-defined functions.); imputing values to a portion of the receipt-level transaction data (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which allows for a random forest classification model that predicts ‘food business category’ based on real-time snapshots and aggregated data features.); removing outlier values from the receipt-level transaction data (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which discloses reclassification of sellers can be accomplished by a random forest classification model that predicts food business category based on real-time snapshots and aggregated data features obtained from payment activity.); or applying price constraints or basket constraints to receipt-level transaction data (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which discloses reclassification of sellers can be accomplished by a random forest classification model that predicts food business category based on real-time snapshots and aggregated data features obtained from payment activity.). As per claim 8, Brosamer teaches the system of claim 5, wherein: the first attribute data includes at least one of a location of the first store, population metrics associated with the first store, a sector category associated with the first store, or transaction trend information associated with the first store (e.g. Brosamer, see column 19, lines 1-28, which discloses geo-location data includes location of payment hardware, merchants, buyers, and other associated entities.); and the second attribute data includes at least one of a location of the at least one second store, population metrics associated with the at least one second store, a sector category associated with the at least one second store, or transaction trend information associated with the at least one second store (e.g. Brosamer, see column 19, lines 1-28, which discloses geo-location data includes location of payment hardware, merchants, buyers, and other associated entities.). As per claim 9, Brosamer teaches the system of claim 8, the operations further comprising extrapolating, by the localized machine learning model or another machine learning model, the sector category associated with the first store or the sector category associated with the at least one second store, based on a correlation learned from receipt-level transaction data (e.g. Brosamer, see column 19, lines 29-36, which discloses the timing and sequential patterns contain a wealth of information about a business ranging from types of business and operating hours to busy times and seasonality.). As per claim 10, Brosamer teaches the system of claim 8, wherein: the first attribute data includes first transaction trend information associated with the first store the second attribute data includes second transaction trend information associated with the at least one second store; and the first transaction trend information and the second transaction trend information both include at least one of a Stock Keeping Unit (SKU) number, an item purchase amount, a payment method identifier, a purchase time, or a discount identifier (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which discloses reclassification of sellers can be accomplished by a random forest classification model that predicts food business category based on real-time snapshots and aggregated data features obtained from payment activity.). As per claim 11, Brosamer teaches the system of claim 10, wherein applying the localized machine learning model to the model input generates at least one predicted brand associated with the transaction (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which discloses reclassification of sellers can be accomplished by a random forest classification model that predicts food business category based on real-time snapshots and aggregated data features obtained from payment activity.). As per claim 12, Brosamer teaches the system of claim 1, wherein applying the localized machine learning model to the model input further generates predicted sub-categories of items associated with the transaction (e.g. Brosamer, see column 24, lines 43-56 and Figure 10, which discloses reclassification of sellers can be accomplished by a random forest classification model that predicts food business category based on real-time snapshots and aggregated data features obtained from payment activity.). As per claim 13, Brosamer teaches the system of claim 1, wherein the operations further comprise generating a recommendation based on the generated predicted categories of items associated with the transaction (e.g. Brosamer, see column 16,l ines 18-33 and Figure 2, which discloses the payment processing system may optionally send class recommendations to the POS device of the merchant.). As per claim 14, Brosamer teaches the system of claim 1, wherein the operations further comprise causing the alteration of, based on the generated predicted categories of items associated with the transaction, a graphic presented at a display (e.g. Brosamer, see column23, lines 39-55 and Figure 9, which discloses the display may have a touch sensor associated with the display to provide a touchscreen display configured to receive touch inputs for enabling interaction with a graphical user interface.). As per claim 15, Brosamer teaches the system of claim 1, wherein the operations further comprise: generating, based on the generated predicted categories of items associated with the transaction, an offer notification (e.g. Brosamer, see column 14, lines 15-32 and Figure 1, which discloses the payment processing system can use the reviews of the merchant to determine which products the merchant is actually providing to customers. For example, the merchant may report to the payment processing system that the merchant offers electronics and food in inventory to customers, which may cause the merchant to be placed with a first class.); and transmitting the offer notification to a user device (e.g. Brosamer, see column 14, lines 15-32 and Figure 1, which discloses the payment processing system can use the reviews of the merchant to determine which products the merchant is actually providing to customers. For example, the merchant may report to the payment processing system that the merchant offers electronics and food in inventory to customers, which may cause the merchant to be placed with a first class.). Conclusion The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure. See attached PTO-892 that includes additional prior art of record describing the general state of the art in which the invention is directed to. THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a). A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action. Contact Information Any inquiry concerning this communication or earlier communications from the examiner should be directed to FARHAN M SYED whose telephone number is (571)272-7191. The examiner can normally be reached M-F 8:30AM-5:30PM. 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, Apu Mofiz can be reached at 571-272-4080. 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. /FARHAN M SYED/Primary Examiner, Art Unit 2161 March 26, 2026
Read full office action

Prosecution Timeline

Dec 29, 2022
Application Filed
Sep 09, 2025
Non-Final Rejection — §102
Jan 09, 2026
Response Filed
Mar 26, 2026
Final Rejection — §102 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12602674
SYSTEM AND METHOD FOR REMOTELY MODIFYING CLOUD-BASED CLIENT APPLICATIONS
2y 5m to grant Granted Apr 14, 2026
Patent 12597071
SYSTEMS AND METHODS OF SWITCHING BETWEEN A FIRST EXECUTION MODE AND SECOND EXECUTION MODE FOR MATCH PROCESSING
2y 5m to grant Granted Apr 07, 2026
Patent 12591792
DYNAMICALLY ENRICHING SHARED KNOWLEDGE GRAPHS
2y 5m to grant Granted Mar 31, 2026
Patent 12579470
SYSTEMS AND METHODS FOR DISTRIBUTING LAYERS OF SPECIAL MIXTURE-OF-EXPERTS MACHINE LEARNING MODELS
2y 5m to grant Granted Mar 17, 2026
Patent 12579207
LOCATION-BASED SEARCHING USING A SEARCH AREA THAT CORRESPONDS TO A GEOGRAPHICAL LOCATION OF A COMPUTING DEVICE
2y 5m to grant Granted Mar 17, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

AI Strategy Recommendation

Get an AI-powered prosecution strategy using examiner precedents, rejection analysis, and claim mapping.
Powered by AI — typically takes 5-10 seconds

Prosecution Projections

3-4
Expected OA Rounds
75%
Grant Probability
98%
With Interview (+23.4%)
3y 9m
Median Time to Grant
Moderate
PTA Risk
Based on 829 resolved cases by this examiner. Grant probability derived from career allow rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month