System and method for leveraging pre-trained embeddings to detect and merge merchant data

Abstract

A system including one or more processors and a computer-readable, non-transitory medium including instructions which cause at least one of the one or more processors to obtain merchant data including a plurality of merchants, obtain a set of word embeddings extracted using a large language model, refine the set of word embeddings by executing a machine-learning model using as input the merchant data to obtain a set of merchant embeddings, determine a first cluster of first merchant embeddings and a second cluster of second merchant embeddings within the set of merchant embeddings, determine a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the second embeddings, and merge the first cluster and the second cluster based on a similarity of the first name and the second name to obtain a merged cluster.

Description

(19) *EP004647998A1* (11) EP 4 647 998 A1 (12) EUROPEAN PATENT APPLICATION (43) Date of publication: 12.11.2025 Bulletin 2025 / 46 (21) Application number: 25171818.5 (22) Date of filing: 22.04.2025 (51) International Patent Classification (IPC): G06Q 30 / 06 (2023.01) G06F 16 / 00 (2019.01) (52) Cooperative Patent Classification (CPC): G06Q 30 / 06; G06F 16 / 906; G06F 18 / 22; G06F 40 / 20; G06N 20 / 00 (84) Designated Contracting States: AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC ME MK MT NL NO PL PT RO RS SE SI SK SM TR Designated Extension States: BA Designated Validation States: GE KH LA MA MD TN (30) Priority: 10.05.2024 US 202463645404 P (71) Applicant: Fiserv, Inc. Milwaukee, WI 53203 (US) (72) Inventors: • ZHANG, Zhiqiang San Ramon (US) • TANG, Yi Fremont (US) • JIANG, Xinshu Sunnyvale (US) (74) Representative: Haseltine Lake Kempner LLP One Portwall Square Portwall Lane Bristol BS1 6BH (GB) (54) SYSTEM AND METHOD FOR LEVERAGING PRE‑TRAINED EMBEDDINGS TO DETECT AND MERGE MERCHANT DATA (57) A system including one or more processors and a computer-readable, non-transitory medium including instructions which cause at least one of the one or more processors to obtain merchant data including a plurality ofmerchants, obtain a set of word embeddings extracted using a large language model, refine the set of word embeddings by executing a machine-learning model using as input the merchant data to obtain a set of merchant embeddings, determine a first cluster of first merchant embeddings and a second cluster of second merchant embeddingswithin the set ofmerchant embed- dings, determine a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the second embeddings, andmerge the first cluster and the secondcluster basedonasimilarity of the first name and the second name to obtain a merged cluster. EP 4 64 7 99 8 A 1 Processed by Luminess, 75001 PARIS (FR) 2 1 EP 4 647 998 A1 2 Description CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to US Provi- sional Application No. 63 / 645,404, filed May 10, 2024, which application is incorporated herein by reference. BACKGROUND

[0002] Stores of a merchant may be identified in trans- action data differently, causing the merchant stores to be incorrectly identified as stores associated with different merchants, causing confusion in identifying parties to a transaction. SUMMARY

[0003] Various aspects of the disclosure may now be described with regard to certain examples and embodi- ments, which are intended to illustrate but not limit the disclosure. Although the examples and embodiments described herein may focus on, for the purpose of illus- tration, specific systemsandprocesses, one of skill in the artmay appreciate the examples are illustrative only, and are not intended to be limiting.

[0004] Aspects of the present disclosure relate to a system including one or more processors, and a compu- ter-readable, non-transitory medium including instruc- tions which, when executed by the one or more proces- sors, cause at least one of the one or more processors to obtain merchant data including a plurality of merchant identifiers, obtain a set of word embeddings extracted using a large language model, refine the set of word embeddings by executing a machine-learning model using as input the merchant data to obtain a set of merchant embeddings, determine a first cluster of first merchant embeddings and a second cluster of second merchant embeddingswithin the set ofmerchant embed- dings, determine a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the second embeddings, andmerge the first cluster and the secondcluster basedonasimilarity of the first name and the second name to obtain a merged cluster, the merged cluster corresponding to a merchant identifier of the plurality of merchant identifiers.

[0005] In some implementations, refining the set of word embeddings includes generating, by the ma- chine-learningmodel, a predicted category for eachword embedding of the set of word embeddings, and refining thesetofwordembeddingsbasedonacomparisonof the predicted category for each word embedding and a cor- responding category label in the merchant data. In some implementations, refining the set of word embeddings includes determining adistancebetweenafirstmerchant embedding and a second merchant embedding, and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding. In some implementations, determining the first name for thefirst cluster includesdetermining thefirst name for the first cluster based on a frequency of words within the first embeddings. In some implementations, determining the first name for the first cluster includes validating the first name based on comparing the first name to a set of merchant names. In some implementa- tions, determining the first name for the first cluster includes determining the set of merchant names based on additional data. In some implementations, the instruc- tions further cause the one or more processors to modify the first name based on a similarity comparison between the first name and a merchant name and the set of merchant names.

[0006] Aspects of the present disclosure are directed to a method including obtaining merchant data including a plurality of merchants, obtaining a set of word embed- dings extracted using a large language model, refining the set of word embeddings by executing a machine- learningmodel using as input themerchant data to obtain a set ofmerchant embeddings, determining a first cluster of first merchant embeddings and a second cluster of secondmerchant embeddingswithin the set ofmerchant embeddings, determining a first name for the first cluster basedon the first embeddings anda secondname for the second cluster based on the second embeddings, and merging the first cluster and the second cluster based on a similarity of the first name and the second name to obtain a merged cluster.

[0007] In some implementations, refining the set of word embeddings includes generating, by the ma- chine-learningmodel, a predicted category for eachword embedding of the set of word embeddings, and refining thesetofwordembeddingsbasedonacomparisonof the predicted category for each word embedding and a cor- responding category label in the merchant data. In some implementations, refining the set of word embeddings includes determining adistancebetweenafirstmerchant embedding and a second merchant embedding, and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding. In some implementations, determining the first name for thefirst cluster includesdetermining thefirst name for the first cluster based on a frequency of words within the first embeddings. In some implementations, determining the first name for the first cluster includes validating the first name based on comparing the first name to a set of merchant names. In some implementa- tions, determining the first name for the first cluster includes determining the set of merchant names based on additional data. In some implementations, the instruc- tions further cause the one or more processors to modify the first name based on a similarity comparison between the first name and a merchant name and the set of merchant names. In some implementations, the method includes modifying the first name based on a similarity 5 10 15 20 25 30 35 40 45 50 55 3 3 EP 4 647 998 A1 4 comparison between the first name and a merchant name of the set of merchant names.

[0008] Aspects of the present disclosure are directed toacomputer-readable, non-transitorymedium including instructions which, when executed by one or more pro- cessors, causeat least oneof theoneormoreprocessors to obtain merchant data including a plurality of mer- chants, obtain a set of word embeddings extracted using a large language model, refine the set of word embed- dings by executing a machine-learning model using as input the merchant data to obtain a set of merchant embeddings, determine a first cluster of first merchant embeddings and a second cluster of second merchant embeddings within the set of merchant embeddings, determine a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the second embeddings, and merge the first cluster and the second cluster based on a simi- larity of the first name and the second name to obtain a merged cluster.

[0009] In some implementations, refining the set of word embeddings includes generating, by the ma- chine-learningmodel, a predicted category for eachword embedding of the set of word embeddings, and refining thesetofwordembeddingsbasedonacomparisonof the predicted category for each word embedding and a cor- responding category label in the merchant data. In some implementations, refining the set of word embeddings includes determining adistancebetweenafirstmerchant embedding and a second merchant embedding, and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding. In some implementations, determining the first name for thefirst cluster includesdetermining thefirst name for the first cluster based on a frequency of words within the first embeddings. In some implementations, determining the first name for the first cluster includes validating the first name based on comparing the first name to a set of merchant names. In some implementa- tions, determining the first name for the first cluster includes determining the set of merchant names based on additional data. In some implementations, the instruc- tions further cause the one or more processors to modify the first name based on a similarity comparison between the first name and a merchant name and the set of merchant names. In some implementations, the instruc- tions further cause the one or more processors to modify the first name based on a similarity comparison between the first name and a merchant name of the set of mer- chant names.

[0010] The foregoingsummary is illustrativeonlyand is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features de- scribed above, further aspects, embodiments, and fea- turesmay become apparent by reference to the following drawings and the detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] FIG. 1 is an example block diagram of a system for identifying and aggregating merchant stores in mer- chant data. FIG. 2 is an example block diagram illustrating de- tails of the embeddings tuning engine of FIG. 1. FIG. 3 is an example block diagram illustrating de- tails of the stage‑1 fine tuning of FIG. 2. FIG. 4 is an example block diagram illustrating de- tails of the stage‑2 fine tuning of FIG. 2. FIG. 5 is an example block diagram of a system for clustering merchant data. FIG. 6 is an example block diagram of a system for generating anchor names. FIG. 7 is an example block diagram illustrating how the clustering engine of FIG. 5 generates cluster names. FIG. 8 is an example flow diagram of a method for merging clusters. FIG. 9 is an example flow diagram of a method for merging clusters and noise data. FIG. 10 is an example block diagram of a computing system 100.

[0012] The foregoing and other features of the present disclosure may become apparent from the following de- scription and appended claims, taken in conjunction with the accompanying drawings. Understanding that these drawings depict only several embodiments in accor- dance with the disclosure and are therefore, not to be considered limiting of its scope, the disclosure may be described with additional specificity and detail through use of the accompanying drawings. DETAILED DESCRIPTION

[0013] In the following detailed description, reference ismade to theaccompanying drawings,which formapart hereof. In the drawings, similar symbols typically identify similar components, unless context dictates otherwise. The illustrative embodiments described in the detailed description, drawings, and claims are not meant to be limiting. Other embodiments may be utilized, and other changes may be made, without departing from the spirit or scope of the subject matter presented here. It may be readily understood that the aspects of the present dis- closure, as generally described herein, and illustrated in the figures,maybearranged, substituted, combined, and designed in a wide variety of different configurations, all of which are explicitly contemplated andmade part of this disclosure.

[0014] Aspects of the present disclosure relate to ac- curately and automatically identifying and aggregating stores associated with amerchant. Transaction data and store descriptors generally describe stores differently, 5 10 15 20 25 30 35 40 45 50 55 4 5 EP 4 647 998 A1 6 even if theybelong to the samemerchant. This cancause confusion when customers review past transactions and try to identify real and fraudulent transactions. For ex- ample, if a customer made a purchase at a WALMART store, the customer may be confused by a transaction description which describes the store based on its ad- dress. Similarly, attempts to analyze spending habits and trends may be frustrated by inaccurate or incomplete mappings of stores tomerchants. The present disclosure provides for using machine-learning models to automa- tically and accurately identify and aggregate merchant stores. The machine-learning models and processes may leverage pre-trained word embeddings to reduce a cost and complexity of the training process. The pre- trained word embeddings can be refined to more accu- rately reflect merchant attributes. In this way, existing word embeddings that have been previously trained can be refined and adapted for purposes of identifying and aggregating different stores belonging to the same mer- chant.

[0015] While various examples and embodiments herein discuss aggregating merchant stores, the sys- tems and methods discussed herein are applicable to identifying and aggregating other kinds of data. Lever- aging pre-trained word embeddings may be advanta- geous in identifying and aggregating various kinds of text-based data. In an example, aspects of the present disclosuremay be applied to identifying and aggregating data associated with a person across several online accounts.

[0016] FIG. 1 is an example block diagram of a system 100 for identifying and aggregating merchant stores in merchant data. The system 100 receives as input raw merchant data 110 and outputs aggregated merchant data 150. The aggregated merchant data 150 may be obtained by modifying the raw merchant data 110 using the system 100. The rawmerchant data 110may include merchant stores and merchants. The rawmerchant data 110 may include merchant stores which are not asso- ciated with their corresponding merchant in the rawmer- chant data 110 and / or merchant stores which are asso- ciatedwithan incorrectmerchant.The rawmerchantdata 110 may include multiple merchant identifiers which cor- respond to a single ground truthmerchant. Thus, the raw merchant data 110 may inaccurately include more mer- chants than ground truth. The aggregatedmerchant data 150may have themultiplemerchant identifiers of the raw merchant data 110 accurately aggregated under a single merchant identifier corresponding to the ground truth merchant.

[0017] The system 100 may include an embeddings tuningengine120, a clusteringengine130, andanormal- ization engine 140. The embeddings tuning engine 120 may use the raw merchant data to fine-tune previously- trained word embeddings to adapt the word embeddings to the task of identifying and aggregating merchant data. The word embeddings may be previously trained / ex- tracted by a large language model, such as BERT. The word embeddingsmay begeneralizedword embeddings trained to represent features of various words. A word embedding for a merchant name may represent ex- tracted features of the merchant name, as extracted by the large language model. The embeddings tuning en- gine 120 may refine the word embedding for the mer- chant name toadapt thewordembedding foraggregating merchant data while leveraging the previous training of thewordembedding.Theembeddings tuningengine120 may allow for transfer learning of the word embedding, with fine-tuning to adapt the word embedding to the merchant data domain. In an example, the embeddings tuning engine 120 fine-tunes a word embedding for the word "Walmart" to obtain a merchant embedding (fine- tuned usingmerchant data) for themerchantWALMART. In this example, the initial embedding for the word "Wal- mart" captures multiple features that are useful for ag- gregating merchant data, and the word embedding is fine-tuned to increase an accuracy of merchant data aggregation. In this way, the process of generating em- beddings for merchant names is greatly shortened and simplified relative to generating embeddings from scratch. The embeddings tuning engine 120may include a multi-stage deep learning model for fine-tuning the word embedding to adapt the word embedding to the merchant domain.

[0018] The clustering engine 130 generates clusters of the fine-tuned word embeddings, also referred to as merchant embeddings, or merchant name embeddings, generated by the embeddings tuning engine 120. The clustering engine 130may include a clustering algorithm. The clustering engine 130 may include a density based clustering algorithm such as Dbscan.

[0019] The normalization engine 140 extracts a nor- malized name for each cluster generated by the cluster- ing engine 130. The normalization engine 140 may gen- erate a normalized name for each cluster based on the merchant embeddings in each cluster. In anexample, the normalization engine 140 may generate the normalized name based on a most frequent merchant name in the embeddings in the cluster. The normalization engine 140 may validate the normalized name for the cluster using multiple verification methods. The normalization engine 140may compare the normalized name for the cluster to a set of anchor names generated based on merchant characteristics, transaction data, the raw merchant data 110, and / or known merchant names. The normalization engine 140 may compare a category of a merchant associated with the normalized name with categories represented in the embeddings in the cluster. In an ex- ample, the normalization engine 140 verifies the normal- ized name by comparing a merchant category code (MCC) associated with a merchant corresponding to the normalized name with a most common MCC repre- sented in the embeddings in the cluster. The normal- ization may compare a URL associated with a merchant corresponding to the normalized namewith aURL froma third-party database which associates URLs with mer- 5 10 15 20 25 30 35 40 45 50 55 5 7 EP 4 647 998 A1 8 chant names.

[0020] In some implementations, the normalization en- gine 140 merges clusters based on the normalized names. The normalization engine 140 may merge clus- ters based on the clusters having a same normalized name, or the clusters having normalized names corre- sponding toasamemerchant. In anexample, thenormal- ization engine 140merges two clusters which each have the normalized name of "Walmart." In an example, the normalization engine 140 merges a first cluster having a normalized name of "Walmart Vision Center" and a sec- ond cluster having a normalized name of "Walmart Vi- sion" which both correspond to an anchor name of "Wal- martVision&Glasses." In this example, theanchor name may correspond to an actual merchant name.

[0021] FIG. 2 is an example block diagram illustrating details of the embeddings tuning engine 120 of FIG. 1. The embeddings tuning engine 120may receive as input a word embedding 222 and output a merchant name embedding 228. The embeddings tuning engine 120 may adapt the word embedding 222 to the merchant domain, refining the word embedding 222 to obtain the merchant name embedding 228. The word embedding 222may be a pre-trainedword embedding, as discussed herein. The word embedding 222may be a word embed- ding of a merchant name trained by a generalized large language model. The merchant name embedding 228 may be obtained by refining the word embedding 222 in stage‑1 fine tuning 224 and stage‑2 fine tuning 226. The embeddings tuning engine 120 may perform the stage‑1 fine tuning 224 and the stage‑2 fine tuning 226 using one or more deep learning machine-learning models.

[0022] The stage‑1 fine tuning 224 may encode mer- chant-specific meaning into each word embedding. The stage‑1fine tuning224mayusemerchant data toencode the merchant-specific meaning into each word embed- ding. The stage‑1 fine tuning 224may include providing a set of merchant names from a merchant database into a machine-learning model as input and using correspond- ing merchant categories as a classification task to fine- tune the word embeddings.

[0023] The stage‑2 fine tuning 226may include provid- ing trainingpairs ofmerchant names to train themachine- learning model used in the stage‑1 fine tuning 224 to recognize merchant name patterns and relationships. The stage‑2 fine tuning 226 may include determining a distance between merchant names in a training pair and applying a loss function to fine-tune the embeddings. In an example, each training pair includes two merchant names and a labeled similarity score which is compared to a generated similarity score representing a distance between two embeddings generated using the two mer- chant names. In this example, the embeddings are fine- tuned to correspond to reduce a difference between the predicted distance and the labeled distance.

[0024] FIG. 3 is an example block diagram illustrating details of the stage‑1 fine tuning 224 of FIG. 2. The stage‑1fine tuning224may includeprovidingamerchant name 310 to a language model 320 to generate an embedding 330 representing features of the merchant corresponding to the merchant name 310. The embed- ding 330 may be initialized using the word embedding 222 of FIG. 2. Generating the embedding 330 using the language model may include refining the word embed- ding 222 of FIG. 2 by executing the language model 320 using as input the merchant name 310. A classification layer 340 may generate classification results 350 using as input the embedding 330. In some implementations, the classification results 350 include a predicted mer- chant category for the merchant corresponding to the merchant name 310 based on the embedding 330. The predicted merchant category may be compared to a labeled merchant category of the merchant name 310 to update the embedding 330. In an example, the classi- fication layer 340 is executed using as input the embed- ding 330 and generates a predicted MCC for the mer- chant name 310 which is compared to a labeledMCC for the merchant name 310. Based on the comparison be- tween the predicted MCC and the labeled MCC, the embedding 330 is updated or refined. In some imple- mentations, the classification results 350 are used to update the language model 320 such that the language model 320 generates the refined embedding when exe- cuted using themerchant name 310 as input. In this way, the embedding is refined by refining the language model 320 such that the embedding generated by the language model 320 based on the merchant name 310 reflects the context of the merchant domain. In this way, the stage‑1 fine tuning 224 may leverage a pre-trained word embed- ding andapre-trained languagemodel, reducing the cost and complexity of generating merchant embeddings. In some implementations, the stage‑1 fine tuning 224 is a deep learning model including the classification layer 340.

[0025] FIG. 4 is an example block diagram illustrating details of the stage‑2 fine tuning 226 of FIG. 2. In some implementations, the stage‑2fine tuning226may include retrofitting the refined embedding 330 of FIG. 3 to be further adapted to themerchant domain. The stage‑2 fine tuning 226 may include providing training pairs to reflect merchant name patterns and relationships. In some im- plementations, the training pairs each include or are associated with labeled similarity values reflecting a si- milarity of the training pair or labeled distances reflecting a distance between the training pair. The stage‑2 fine tuning 226 may include providing a first merchant name 410 and a second merchant name 420 as input to the languagemodel 320ofFIG.3 togenerateafirstmerchant name embedding 412 and a second merchant name embedding 422. The first merchant name 410 and the second merchant name 420 may be a training pair. The language model 320 may be updated in the stage‑1 fine tuning 224 such that the first merchant name embedding 412 and the second merchant name embedding 422 are refined based on the merchant domain. A similarity func- tion layer 430 receives as input the first merchant name 5 10 15 20 25 30 35 40 45 50 55 6 9 EP 4 647 998 A1 10 embedding 412 and the second merchant name embed- ding 422 and determines a distance between the first merchant name embedding 412 and the second mer- chant name embedding 422. In some implementations, the similarity function layer 430 includes an absolute value distance function and / or a dot product function which calculates a dot product of the first merchant name embedding 412 and the second merchant name embed- ding 422.

[0026] A regression layer 440 may generate a result 450 or similarity score for the first merchant name em- bedding 412 and the secondmerchant name embedding 422 based on the distance determined by the similarity function layer 430. In some implementations, the regres- sion layer 440 uses regression loss or a loss function to measure the distance between the first merchant name embedding 412 and the second merchant name embed- ding 422. The result 450may be compared to the labeled similarity value or labeled distance of the training pair to update the languagemodel 320. In thisway, the language model 320 is updated to reflect the labeled similarity or labeled distance between the first merchant name 410 and the second merchant name 420, further refining the embeddings generated by the language model 320.

[0027] In an example, a training pair of "Walmart" and "B-Mart" may have a labeled distance of "10," represent- ing a lack of relationship or similarity. In an example, a training pair of "Walmart" and "Walmart San Jose" may have a labeled distance of "1," representing a strong relationship or similarity. In an example, a training pair of "Walmart" and "Walmart Pharmacy" may have a la- beled distance of "2," representing a strong relationship or similarity. In an example, a training pair of "Walmart Vision" and "Walmart Pharmacy" may have a labeled distance of "3," representing a strong relationship or similarity. The language model 320 may be executed using a plurality of training pairs in the stage‑2 fine tuning 226. The languagemodel 320may be updated such that the result 450 approaches and / or matches the labeled similarity values or labeled distances for the plurality of training pairs. In some implementations, a loss function is applied to the language model 320 using the result 450 and the labeled similarity values or labeled distances as input to reduce a difference between the determined distance and the labeled similarity values or labeled distances. In this way, the language model 320 learns relationships and similarities between merchant names in order to understand how different merchant names represent different relationships. For example, the lan- guage model 320 may learn that adding a location name to a merchant name signifies that a store belongs to a merchant and is located in the location. In this example, the language model 320 may learn that "Walmart San Jose" is a store associatedwith themerchantWALMART and is located in San Jose.

[0028] FIG. 5 is an example block diagram of a system 500 for clustering merchant data. The system 500 may include a clustering engine 520 and a normalization engine 540. The clustering engine 520 may be similar to or the same as the clustering engine 130 of FIG. 1. The normalization engine 540may be similar to, the same as, or part of the normalization engine 140 of FIG. 1.

[0029] The clustering engine 520may receive as input merchant embeddings 510. The merchant embeddings 510 may be the refined merchant embeddings, or mer- chant name embedding 228 generated by the embed- dings tuning engine 120. The clustering engine 520 may be executed using as input the merchant embeddings 510 to generate clusters of embeddings. The clustering engine 520 may cluster the merchant embeddings 510 based on the features represented in the merchant em- beddings 510 to cluster similar merchant embeddings 510. The clustering engine 520 determines a standar- dized name for each cluster of embeddings. The cluster- ing engine520maydetermine the standardizedname for a cluster based on the embeddings in the cluster. The clustering engine 520 provides the clusters of embed- dings to the normalization engine 540.

[0030] The normalization engine 540 may receive as input the clusters of embeddings from the clustering engine 520 and anchor names 530 associated with mer- chant names to normalize the names of the clusters of embeddings. In some implementations, the anchor names 530 are determined separate from the names of the clusters determined by the clustering engine 520 in order to verify and normalize the names of the clusters. The normalization engine 540may, based on the normal- ized namesof the clusters,merge oneormore clusters of embeddings. A result 550 of the normalization engine 540 may include a set of clusters, including merged clusters, each having a normalized name. The set of clusters may represent relationships or similarities be- tween merchants. In an example, a cluster of embed- dingsmay represent a set ofmerchants corresponding to a single ground truth merchant. In this way, transactions of the set of merchants can be accurately determined to be associated with the single ground truth merchant.

[0031] FIG. 6 is an example block diagram of a system 600 for generating anchor names the anchor names 530 of FIG. 5. The system 600 may include a first filtering engine 612 which filters payment terminal data 610. The payment terminal data 610 may include data from pay- ment terminals, suchaspoint-of-sale (POS)devices.The payment terminal data 610 may include merchant names, merchant categories, and terminal counts. In an example, the payment terminal data 610 includes, for eachmerchant name, a number of payment terminals associated with each merchant category. In an example, the payment terminal data 610 includes a count of pay- ment terminals associated with each MCC for the mer- chant names "Walmart," "Walmart Supercenter," and "Walmart Store." The first filtering engine 612 may apply knee point filtering to determine a set of names from the payment terminal data 610. In an example, the first filtering engine 612may identify merchant names having payment terminal counts per merchant category which 5 10 15 20 25 30 35 40 45 50 55 7 11 EP 4 647 998 A1 12 includeakneeanddetermine that the identifiedmerchant names are valid merchant names. The first filtering en- gine 612 may output a first set of merchant names.

[0032] The system 600 may include a second filtering engine 622 which filters store data 620 received from merchant stores. The store data 620 may include mer- chant names, merchant domain names, merchant cate- gories, and / or merchant store counts. In an example, the storedata620 includes, for eachmerchant name,acount of stores associated with each MCC. In an example, the store data 620 includes a chart showing store count per MCC for each of the merchant names of "Walmart," "Walmart Bakery," "Walmart Vision&Glasses," "Walmart Supercenter," "Walmart Distribution Center," "Walmart Neighborhood Market," and "Walmart Grocery Pickup & Delivery." The second filtering engine 622 may apply wrong domain filtering and knee point filtering to the store data 620.

[0033] The second filtering engine 622 may apply the wrong domain filtering by comparing a merchant’s mer- chant domain name (e.g., merchant website URL) to the merchant’s merchant name to determine whether the merchant domain name from the store data 620 corre- sponds to themerchant name from the store data 620. In some implementations, the merchant domain name and the merchant name are extracted from a website of the merchant (web-scrapeddata). In some implementations, comparing the merchant domain name to the store data includes verifying one or more of the merchant domain name and the merchant name. In an example, verifying the merchant domain name may include verifying, using other data, that themerchant domain name is associated with the merchant name.

[0034] The second filtering engine 622 may apply the knee point filtering to determine a set of names from the store data 620. In an example, the second filtering engine 622 may identify merchant names having store counts per merchant category which include a knee and deter- mine that the identified merchant names are valid mer- chant names. In an example, the second filtering engine 622 analyzes charts showing store count per MCC to identify the knee. The second filtering engine 622 may output a second set of merchant names.

[0035] The system 600 may include a third filtering engine 632 which filters merchant data 630. The mer- chant data 630 may include merchant names, merchant domains, and merchant rankings. The merchant data 630maybeobtained frommerchant systemsand / or from third-party systems. The third filtering engine 632 may apply wrong name filtering and ranking filtering. The third filteringengine632mayapply thewrongnamefilteringby comparing themerchant names to themerchant domain, the merchant ranking, and / or additional data associated with the merchant names. The third filtering engine 632 may apply the ranking filtering by comparing the mer- chant ranking to a ranking of merchants. In an example, third filtering engine 632 compares a size rank of a merchant from the merchant data 630 with a ranking of merchants by size to verify the size rank and the associa- tion between the merchant and the size rank. The third filtering engine 632 may output a third set of merchant names.

[0036] The anchor names 530may include the first set of merchant names, the second set of merchant names, and / or the third set of merchant names. In some imple- mentations, the anchor names 530 include merchant names that are present in two or more of the first set of merchant names, the second set of merchant names, and the third set of merchant names. In some implemen- tations, the anchor names 530 include merchant names thatarepresent ineachof thefirst setofmerchantnames, the second set of merchant names, and the third set of merchant names. In this way, the anchor names 530 include a set of merchant names which are extracted from multiple sources of data to verify that the set of merchant names correspond to ground truth merchants.

[0037] In some implementations, the anchor names 530 are determined by the normalization engine 540 of FIG. 5 and the normalization engine 540 includes the system600. In some implementations, theanchornames 530 are used by the normalization engine 540 of FIG. 5 to generate the result 550 of FIG. 5.

[0038] FIG. 7 is an example block diagram illustrating how the clustering engine 520 of FIG. 5 generates cluster names. The clustering engine 520 may take as input merchant embeddings, such as the merchant embed- dings 510 of FIG. 5 and generate a plurality of clusters. In an example, the clustering engine 520 utilizes a density- based clustering algorithm such as DBscan to generate the plurality of cluster. The clustering engine 520 may generate a plurality of clusters including a first cluster 710, a second cluster 720, and an nth cluster 730. The clustering engine 520 may generate a noise cluster 740 including embeddings which are not included in the plur- ality of clusters. The clustering engine 520 may deter- mine a set of n-grams for each cluster in the plurality of clusters. In some implementations, the set of n-grams for each cluster may be word-grams. The clustering engine 520 may perform filtering on the set of n-grams for each cluster to determine a name for each cluster. In some implementations, the filtering includes knee point filtering based on a frequency of n-grams within the set of n- grams.

[0039] The clustering engine 520 may determine a set of first n-grams 712 for the first cluster 710 based on the embeddings in the first cluster. In an example, the clus- tering engine 520 determines that the first cluster 710 includes sixty unique store names. The set of first n- grams 712 may be a set of word grams from the embed- dings of the first cluster 710. The clustering engine 520 may perform first filtering 714 on the set of first n-grams 712 to determine a first name 716 for the first cluster 710. The first filtering 714 may include knee point filtering. In an example, the set of first n-grams 712 includes sixty word grams of "walmart," fifty-six word grams of "wal- mart, vision," fifty-three word grams of "walmart, vision, 5 10 15 20 25 30 35 40 45 50 55 8 13 EP 4 647 998 A1 14 center," and three word grams of "walmart, vision, and, glasses." In this example, the first filtering 714 may determine that a knee exists at the n-grams of "walmart, vision, center," causing the first name 716 to be "walmart vision center." Thus, the clustering engine 520 may de- termine the first name 716 for the first cluster 710 based ona frequencyof n-grams, a frequencyofword-grams, or a frequency of words within the embeddings of the first cluster 710.

[0040] Similarly, the clustering engine 520 generates the second cluster 720, extracts a set of second n-grams 722 from the second cluster 720, and applies second filtering 724 to the set of second n-grams 722 to deter- mine a second name 726 for the second cluster 720. Similarly, the clustering engine 520 generates the nth cluster 730, extracts a set of nth n-grams732 from thenth cluster730, andappliesnthfiltering734 to theset of nthn- grams 732 to determine an nth name 736 for the nth cluster 730.

[0041] In some implementations, the first filtering 714, the second filtering 724, and the nth filtering 734 are the same, and / or use the same filtering method (e.g., knee point filtering).

[0042] The clustering engine 520 may identify embed- dings that are not included in the plurality of clusters as noise belonging to the noise cluster 740. The clustering engine 520 may perform category filtering 742 on the noise cluster 740 to determine noise names 744 for the noise cluster 740. The category filtering may include comparing merchant categories, such as MCCs in the embeddings in the noise cluster 740 to identify a most common category. The noise cluster 740 may include multiple different categories which each receive a differ- ent name. In this way, the noise names 744 identify characteristics of the noise.

[0043] FIG. 8 is an example flow diagram of an exam- ple method 800 for merging clusters. The method 800 may include more, fewer, or different operations than illustrated. The operationsmay be performed in the order shown, in a different order, or concurrently.

[0044] At operation 812, a similarity comparison is performed between a first cluster name 810 and a set of anchornames820.Thefirst clustername810maybea cluster name determined by the clustering engine 520 of FIG. 5. The set of anchor names 820 may be the anchor names 530 of FIG. 5 as determined using the system600 of FIG. 6. Performing the similarity comparison may include determining a similarity between the first cluster name 810 and each anchor name of the set of anchor names 820. At operation 814, the determined similarities are compared to a threshold similarity to determine whether a similarity between the first cluster name 810 and an anchor name of the set of anchor names 820 exceeds the threshold similarity. In response to a deter- mined similarity between the first cluster name and an anchor name of the set of anchor names 820 exceeding the threshold similarity, the first cluster name 810 is set to the anchor name at operation 816. In an example, the similarity comparison is performed between a cluster name of "walmart vision center" and a set of anchor names including an anchor name of "walmart vision & glasses," causing the cluster name to be set, or changed to "walmart vision & glasses." In this way, the first cluster name 810 is set to be a name (anchor name) extracted from data independent of the embeddings. This allows the clusters to bemapped to realmerchant names based on the characteristics of the embeddings in the clusters.

[0045] At operation 832, a similarity comparison is performed between a second cluster name 830 and the set of anchor names 820 to determine similarities between the second cluster name 830 and each anchor name of the set of anchor names 820. At operation 834, the determined similarities are compared to a threshold similarity to determine whether a similarity between the second cluster name 830 and an anchor name of the set of anchor names 820 exceeds the threshold similarity. At operation 836, based on a similarity between the second cluster name 830 and an anchor name of the set of anchor names 820 exceeding the threshold similarity, the second cluster name 830 is set to be the anchor name.

[0046] At operation 840, the first cluster name 810, set to the corresponding anchor name, and the second clus- ter name 830, set to the corresponding anchor name, are compared to determine whether the first cluster name 810 and the second cluster name 830match, or whether theanchornames foreachof thefirst andsecondclusters match.At operation850, in response to thecluster names or anchor names matching, the first and second clusters are merged. In some implementations, at operation 840, multiple cluster namesofmultiple clusters are compared. In an example, a plurality of cluster names are examined at operation 840, causing all clusters with matching names to be merged at operation 850. In some imple- mentations, names of merged clusters are compared to names of clusters at operation 840 and merged clusters and clusters are merged at operation 850 based on matching names. In this way, clusters which are mapped to the same merchants are merged. This allows for accurately and automatically merging merchant identi- fiers to correspond to real merchants.

[0047] Clusters may bemerged at various hierarchical levels representing different levels of analysis. In an example, a cluster having a name of "Walmart Bakery" and a cluster having a nameof "Walmart Pharmacy"may be separate at a first hierarchical level and merged at a second, higher hierarchical level into a cluster having a name of "Walmart." In this way, a hierarchy of merchant stores may be constructed to show the relationships between different merchants and to accurately and auto- matically identify merchant stores within the hierarchy of merchant stores. The hierarchy of merchant stores may be used to inform users of spending patterns, inform users of transaction location, and / or to analyze or track spending habits.

[0048] FIG. 9 is an example flow diagram of a method 5 10 15 20 25 30 35 40 45 50 55 9 15 EP 4 647 998 A1 16 900 formerging clusters and noise data. Themethod 900 may include more, fewer, or different operations than illustrated. The operationsmay be performed in the order shown, in a different order, or concurrently.

[0049] At operation 912, a similarity comparison is performed between a cluster name 910 and a set of anchor names 920. The cluster name 910 may be a cluster name determined by the clustering engine 520 ofFIG.5.Thesetof anchor names920maybe theanchor names 530 of FIG. 5 as determined using the system600 of FIG. 6. Performing the similarity comparison may include determining a similarity between the cluster name 910 and each anchor name of the set of anchor names 920. At operation 914, the determined similarities are compared to a threshold similarity to determine whether a similarity between the cluster name 910 and an anchor name of the set of anchor names 920 exceeds the threshold similarity. In response to a determined similarity between the first cluster name and an anchor name of the set of anchor names 920 exceeding the threshold similarity, the cluster name 910 is set to the anchor name at operation 916. In an example, the simi- larity comparison is performedbetweenacluster nameof "walmart vision center" and a set of anchor names in- cluding an anchor name of "walmart vision & glasses," causing the cluster name to be set, or changed to "wal- mart vision&glasses." In thisway, theclustername910 is set to be a name (anchor name) extracted from data independent of the embeddings. This allows the clusters to be mapped to real merchant names based on the characteristics of the embeddings in the clusters.

[0050] At operation 932, a similarity comparison is performed between a noise name 930 and the set of anchor names 920 to determine similarities between the noise name 930 and each anchor name of the set of anchor names 920. At operation 934, the determined similarities are compared to a threshold similarity to determine whether a similarity between the noise name 930 and an anchor name of the set of anchor names 920 exceeds the threshold similarity. At operation 936, based on a similarity between the noise name 930 and an anchor name of the set of anchor names 920 exceeding the threshold similarity, the noise name 930 is set to be the anchor name.

[0051] At operation 940, the cluster name 910, set to the corresponding anchor name, and the noise name 930, set to the corresponding anchor name, are com- pared to determinewhether the cluster name910and the noise name 930match, or whether the anchor names for eachof the cluster and thenoisematch. At operation 950, in response to thecluster namesor anchor namesmatch- ing, the cluster and noise are merged. In some imple- mentations, at operation 940, an anchor name of a merged cluster formed by merging two or more clusters is compared to the noise name930 to determinewhether to merge the merged cluster and the noise at operation 950. In thisway, noise,whichwasnot included in acluster may be rejoined into a cluster based onmatching anchor names. This allows for accurately and automaticallymer- ging merchant identifiers to correspond to real mer- chants.

[0052] FIG. 10 is an example block diagram of a com- puting system 1000, in accordance with some embodi- ments of the present disclosure. The computing system 1000 includes a host device 1005 associated with a memory device 1010. The host device 1005 may be configured to receive input from one or more input de- vices 1015 and provide output to one or more output devices1020.Thehost device1005maybeconfigured to communicate with the memory device 1010, the input devices 1015, and the output devices 1020 via appro- priate interfaces or channels 1025A, 1025B, and 1025C, respectively. The computing system 1000 may be imple- mented in a variety of computing devices such as com- puters (e.g., desktop, laptop, etc.), tablets, personal di- gital assistants, mobile devices, wearable computing devices such as smart watches, other handheld or por- table devices, or any other computing unit suitable for performing operations described herein using the host device 1005.

[0053] Further, some or all of the features described in the present disclosure may be implemented on a client device, a server device, or a cloud / distributed computing environment, or a combination thereof. Additionally, un- less otherwise indicated, functions described herein as being performed by a computing device (e.g., the com- puting system 1000) may be implemented by multiple computing devices in a distributed environment, and vice versa.

[0054] The input devices 1015 may include any of a variety of input technologies such as a keyboard, stylus, touch screen, mouse, track ball, keypad, microphone, voice recognition,motion recognition, remote controllers, input ports, one ormore buttons, dials, joysticks, and any other input peripheral that is associated with the host device1005and that allowsanexternal source, suchasa user, computer, or database, to enter information (e.g., data) into the host device and send instructions to the host device 1005. Similarly, the output devices 1020may include a variety of output technologies such as external memories, databases, printers, speakers, displays, mi- crophones, light emitting diodes, headphones, plotters, speech generating devices, video devices, and any other output peripherals that are configured to receive informa- tion (e.g., data) from thehost device1005. The "data" that is either input into the host device 1005 and / or output from the host device may include any of a variety of textual data, graphical data, video data, sound data, position data, combinations thereof, or other types of analog and / or digital data that is suitable for processing using the computing system 1000.

[0055] The host device 1005 may include one or more Central Processing Unit ("CPU") or Graphics Processing Unit ("GPU") cores or processors 1030A‑1030N thatmay be configured to execute instructions for running one or more applications associated with the host device 1005. 5 10 15 20 25 30 35 40 45 50 55 10 17 EP 4 647 998 A1 18 In some embodiments, the instructions and data needed to run the one or more applications may be stored within thememory device 1010. The host device 1005may also be configured to store the results of running the one or more applications within the memory device 1010. One such application on the host device 1005 may include a merchant data aggregation application 1035. The mer- chant data aggregation application 1035 may be exe- cuted by one or more of the CPU / GPU cores 1030A‑1030N. The instructions to execute the merchant data aggregation application 1035 may be stored within the memory device 1010. The merchant data aggrega- tion application 1035 is described in greater detail above and may perform functions such as described in FIGS. 1‑7and / ormethods suchas themethod800of FIG. 8and themethod900ofFIG.9.Thus, thehost device1005may be configured to request the memory device 1010 to perform a variety of operations. For example, the host device 1005 may request the memory device 1010 to read data, write data, update or delete data, and / or per- form management or other operations.

[0056] To facilitate communication with the memory device 1010, the memory device 1010 may include or be associated with a memory controller 1040. Although the memory controller 1040 is shown as being part of the memory device 1010, in some embodiments, the mem- ory controller 1040may instead be part of the host device 1005 or another element of the computing system 1000 and operatively associated with the memory device 1010. The memory controller 1040 may be configured as a logical block or circuitry that receives instructions from the host device 1005 and performs operations in accordance with those instructions. For example, when the execution of the merchant data aggregation applica- tion 1035 is desired, the host device 1005 may send a request to the memory controller 1040. The memory controller 1040may read the instructions associatedwith the merchant data aggregation application 1035 that are stored within the memory device 1010, and send those instructions back to the host device. In some embodi- ments, those instructionsmaybe temporarily storedwith- in a memory on the host device 1005. One or more of the CPU / GPU cores 1030A‑1030N may then execute those instructions by performing one or more operations called for by those instructions of the merchant data aggrega- tion application 1035.

[0057] The memory device 1010 may include one or more memory circuits 1045 that store data and instruc- tions.Thememorycircuits 1045maybeanyofavariety of memory types, including a variety of volatile memories, non-volatile memories, or a combination thereof. For example, in some embodiments, one or more of the memory circuits 1045 or portions thereof may include NANDflashmemory cores. Inotherembodiments, oneor more of thememory circuits 1045 or portions thereofmay includeNORflashmemory cores,StaticRandomAccess Memory (SRAM) cores, Dynamic RandomAccessMem- ory (DRAM) cores, Magnetoresistive Random Access Memory (MRAM) cores, Phase Change Memory (PCM) cores, Resistive Random Access Memory (Re- RAM) cores, 3D XPoint memory cores, ferroelectric ran- dom-access memory (FeRAM) cores, and other types of memory cores that are suitable for usewithin thememory device 1010. In some embodiments, one or more of the memory circuits 1045 or portions thereof may be config- ured as other types of storage class memory ("SCM"). Generally speaking, the memory circuits 1045 may in- cludeany of a variety ofRandomAccessMemory (RAM), Read-Only Memory (ROM), Programmable ROM (PROM), Erasable PROM (EPROM), Electrically EPROM(EEPROM), hard disk drives, flashdrives,mem- ory tapes, cloud memory, or any combination of primary and / or secondary memory that is suitable for performing the operations described herein.

[0058] It is to be understood that only some compo- nents of the computing system 1000 are shown and described in FIG. 10. However, the computing system 1000 may include other components such as various batteries and power sources, networking interfaces, rou- ters, switches, external memory systems, controllers, etc. Generally speaking, the computing system 1000 may include any of a variety of hardware, software, and / or firmware components that are needed or consid- ered desirable in performing the functions described herein. Similarly, the host device 1005, the input devices 1015, the output devices 1020, and the memory device 1010, including the memory controller 1040 and the memory circuits 1045, may include hardware, software, and / or firmware components that are considered neces- sary or desirable in performing the functions described herein. In addition, in certain embodiments, the memory device 1010may integrate someor all of the components of the host device 1005, including, for example, the CPU / GPUcores1030A‑1030N, and theCPU / GPUcores may be configured to execute the merchant data aggre- gation application 1035, as described herein.

[0059] The various illustrative logical blocks, circuits, modules, routines, and algorithm steps described in con- nection with the embodiments disclosed herein can be implemented as electronic hardware, or combinations of electronic hardware and computer software. To clearly illustrate this interchangeability, various illustrative com- ponents, blocks, modules, and steps have been de- scribed above generally in terms of their functionality. Whether such functionality is implemented as hardware, or as software that runs on hardware, depends upon the particular application and design constraints imposed on the overall system. The described functionality can be implemented in varying ways for each particular applica- tion, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosure.

[0060] Moreover, the various illustrative logical blocks and modules described in connection with the embodi- ments disclosed herein can be implemented or per- formed by a machine, such as a general purpose pro- 5 10 15 20 25 30 35 40 45 50 55 11 19 EP 4 647 998 A1 20 cessor device, a digital signal processor (DSP), an ap- plication specific integrated circuit (ASIC), a field pro- grammable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof de- signed to perform the functions described herein. A con- trol processor can synthesize a model for an FPGA. For example, the control processor can synthesize a model for logical programmable gates to implement a tensor array and / or a pixel array. The control channel can synthesize a model to connect the tensor array and / or pixel array on an FPGA, a reconfigurable chip and / or die, and / or the like. A general purpose processor device can be amicroprocessor, but in the alternative, the processor device can be a controller, microcontroller, or state ma- chine, combinations of the same, or the like. A processor device can include electrical circuitry configured to pro- cess computer-executable instructions. In another em- bodiment, a processor device includes an FPGA or other programmable device that performs logic operations without processing computer-executable instructions. A processor device can also be implemented as a com- bination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microproces- sors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technol- ogy, a processor device may also include primarily ana- log components. For example, some or all of the algo- rithms described herein may be implemented in analog circuitry ormixedanaloganddigital circuitry.A computing environment can include any type of computer system, including, but not limited to, a computer system based on amicroprocessor, amainframe computer, a digital signal processor, a portable computing device, a device con- troller, or a computational engine within an appliance, to name a few.

[0061] The elements of a method, process, routine, or algorithm described in connection with the embodiments disclosedherein canbeembodieddirectly inhardware, in a software module executed by a processor device, or in a combination of the two.Asoftwaremodule can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a re- movable disk, a CD-ROM, or any other form of a non- transitory computer-readable storage medium. An ex- emplary storage medium can be coupled to the proces- sor device such that the processor device can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor device. The processor device and the storagemediumcan reside in anASIC.TheASIC can reside in a user terminal. In the alternative, the processor device and the storage medium can reside as discrete components in a user terminal.

[0062] Conditional language used herein, such as, among others, "can," "could," "might," "may," "e.g.," and the like, unless specifically stated otherwise, or otherwise understood within the context as used, is gen- erally intended to convey that certain embodiments in- clude, while other embodiments do not include, certain features, elements and / or steps. Thus, such conditional language is not generally intended to imply that features, elements and / or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without other input or prompting, whether these features, ele- ments and / or steps are included or are to be performed in any particular embodiment. The terms "comprising," "in- cluding," "having," and the like are synonymous and are used inclusively, in an open-ended fashion, and do not exclude additional elements, features, acts, operations, and so forth. Also, the term "or" is used in its inclusive sense (and not in its exclusive sense) so that when used, for example, to connect a list of elements, the term "or" means one, some, or all of the elements in the list.

[0063] While the above detailed description has shown, described, and pointed out novel features as applied to various embodiments, it can be understood that various omissions, substitutions, and changes in the form and details of the devices or algorithms illustrated can be made without departing from the spirit of the disclosure. As can be recognized, certain embodiments described herein can be embodied within a form that does not provide all of the features and benefits set forth herein, as some features can be used or practiced se- parately from others.

[0064] Thehereindescribedsubjectmatter sometimes illustrates different components contained within, or con- nectedwith, different other components. It is to be under- stood that such depicted architectures are merely ex- emplary, and that in fact many other architectures can be implemented which achieve the same functionality. In a conceptual sense, any arrangement of components to achieve the same functionality is effectively "associated" such that the desired functionality is achieved. Hence, any two components herein combined to achieve a par- ticular functionality can be seen as "associated with" eachother such that thedesired functionality is achieved, irrespective of architectures or intermedial components. Likewise, any twocomponents soassociated canalsobe viewed as being "operably connected," or "operably coupled," to each other to achieve the desired function- ality, and any two components capable of being so as- sociated can also be viewed as being "operably coupl- able," to each other to achieve the desired functionality. Specific examples of operably couplable include but are not limited to physically mateable and / or physically inter- acting components and / or wirelessly interactable and / or wirelessly interacting components and / or logically inter- acting and / or logically interactable components.

[0065] With respect to the use of substantially any plural and / or singular terms herein, those having skill in the art can translate from the plural to the singular and / or from the singular to the plural as is appropriate to the context and / or application. The various singular / plur- 5 10 15 20 25 30 35 40 45 50 55 12 21 EP 4 647 998 A1 22 al permutations may be expressly set forth herein for sake of clarity.

[0066] It will be understood by those within the art that, in general, terms used herein, and especially in the appended claims (e.g., bodies of the appended claims) are generally intended as "open" terms (e.g., the term "including" should be interpreted as "including but not limited to," the term "having" should be interpreted as "having at least," the term "includes" should be inter- preted as "includes but is not limited to," etc.). It will be further understood by thosewithin the art that if a specific number of an introduced claim recitation is intended, such an intent will be explicitly recited in the claim, and in theabsenceof such recitationnosuch intent ispresent. For example, as an aid to understanding, the following appended claims may contain usage of the introductory phrases "at least one" and "one or more" to introduce claim recitations. However, the use of such phrases should not be construed to imply that the introduction of a claim recitation by the indefinite articles "a" or "an" limits any particular claim containing such introduced claim recitation to inventions containing only one such recitation, even when the same claim includes the intro- ductory phrases "one or more" or "at least one" and indefinite articles such as "a" or "an" (e.g., "a" and / or "an" should typically be interpreted tomean "at least one" or "one or more"); the same holds true for the use of definite articles used to introduce claim recitations. In addition, even if a specific number of an introduced claim recitation is explicitly recited, those skilled in the art will recognize that such recitation should typically be inter- preted tomean at least the recited number (e.g., the bare recitation of "two recitations," without other modifiers, typically means at least two recitations, or two or more recitations). Furthermore, in those instances where a convention analogous to "at least one of A, B, and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B, and C" would include but not be limited to systems that haveAalone,Balone,Calone, AandB together, AandC together, B and C together, and / or A, B, and C together, etc.). In those instances, where a convention analogous to "at least one of A, B, or C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B, or C" would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.). It will be further under- stood by those within the art that virtually any disjunctive word and / or phrase presenting two or more alternative terms, whether in the description, claims, or drawings, should be understood to contemplate the possibilities of including one of the terms, either of the terms, or both terms. For example, the phrase "A or B" will be under- stood to include the possibilities of "A" or "B" or "A andB." Further, unless otherwise noted, the use of the words "approximate," "about," "around," "substantially," etc., mean plus or minus ten percent.

[0067] The foregoing description of illustrative embo- diments has been presented for purposes of illustration and of description. It is not intended to be exhaustive or limiting with respect to the precise form disclosed, and modifications and variations are possible in light of the above teachings or may be acquired from practice of the disclosed embodiments. It is intended that the scope of the invention be defined by the claims appended hereto and their equivalents. Claims 1. A system comprising: one or more processors; and a computer-readable, non-transitory medium including instructions which, when executed by the one or more processors, cause at least one of the one or more processors to: obtainmerchant data including a plurality of merchants; obtain a set of word embeddings extracted using a large language model; refine the set of word embeddings by ex- ecuting a machine-learning model using as input the merchant data to obtain a set of merchant embeddings; determine a first cluster of first merchant embeddings and a second cluster of sec- ond merchant embeddings within the set of merchant embeddings; determine a first name for the first cluster basedon thefirst embeddingsandasecond name for the second cluster based on the second embeddings; and merge the first cluster and the second clus- ter basedonasimilarity of thefirst nameand the second name to obtain a merged clus- ter. 2. The system of claim 1, wherein refining the set of word embeddings includes: generating, by the machine-learning model, a predicted category for each word embedding of the set of word embeddings; and refining the set of word embeddings based on a comparison of the predicted category for each word embedding and a corresponding category label in the merchant data. 3. The systemof claim1or 2, wherein refining the set of word embeddings includes: 5 10 15 20 25 30 35 40 45 50 55 13 23 EP 4 647 998 A1 24 determining a distance between a firstmerchant embedding and a secondmerchant embedding; and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding. 4. The system of any of the preceding claims, wherein determining the first name for the first cluster in- cludes: determining the first name for the first cluster based on a frequency of words within the first embeddings; validating thefirst namebasedoncomparing the first name to a set of merchant names; or determining the set of merchant names based on additional data. 5. The system of claim 4, wherein: determining the first name for the first cluster includes determining the set ofmerchant names based on the additional data; and the instructions further cause the one or more processors to modify the first name based on a similarity comparison between the first name and a merchant name and the set of merchant names. 6. A method comprising: obtaining merchant data including a plurality of merchants; obtaining a set of word embeddings extracted using a large language model; refining the set of word embeddings by execut- ing amachine-learningmodel using as input the merchant data to obtain a set of merchant em- beddings; determining a first cluster of first merchant em- beddings and a second cluster of second mer- chant embeddings within the set of merchant embeddings; determining a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the sec- ond embeddings; and merging the first cluster and the second cluster based on a similarity of the first name and the second name to obtain a merged cluster. 7. The method of claim 6, wherein refining the set of word embeddings includes: generating, by the machine-learning model, a predicted category for each word embedding of the set of word embeddings; and refining the set of word embeddings based on a comparison of the predicted category for each word embedding and a corresponding category label in the merchant data. 8. Themethodof claim6or7,wherein refining the set of word embeddings includes: determining a distance between a firstmerchant embedding and a secondmerchant embedding; and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding. 9. The method of any of claims 6 to 8, wherein deter- mining the first name for the first cluster includes: determining the first name for the first cluster based on a frequency of words within the first embeddings; validating thefirst namebasedoncomparing the first name to a set of merchant names; or determining the set of merchant names based on additional data. 10. The method of claim 9, wherein: determining the first name for the first cluster includes determining the set ofmerchant names based on the additional data; and themethod further comprises modifying the first namebasedonasimilarity comparisonbetween the first name and amerchant name of the set of merchant names. 11. Acomputer-readable, non-transitorymedium includ- ing instructions which, when executed by one or more processors, cause at least one of the one or more processors to: obtain merchant data including a plurality of merchants; obtain aset ofwordembeddingsextractedusing a large language model; refine the set of word embeddings by executing a machine-learning model using as input the merchant data to obtain a set of merchant em- beddings; determineafirst cluster of firstmerchant embed- dings and a second cluster of second merchant embeddings within the set of merchant embed- dings; determine a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the second embed- 5 10 15 20 25 30 35 40 45 50 55 14 25 EP 4 647 998 A1 26 dings; and merge the first cluster and the second cluster based on a similarity of the first name and the second name to obtain a merged cluster. 12. The computer-readable, non-transitory medium of claim 11, wherein refining the set of word embed- dings includes: generating, by the machine-learning model, a predicted category for each word embedding of the set of word embeddings; and refining the set of word embeddings based on a comparison of the predicted category for each word embedding and a corresponding category label in the merchant data. 13. The computer-readable, non-transitory medium of claim 11 or 12, wherein refining the set of word embeddings includes: determining a distance between a firstmerchant embedding and a secondmerchant embedding; and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding. 14. The computer-readable, non-transitory medium of any of claims 11 to 13, wherein determining the first name for the first cluster includes: determining the first name for the first cluster based on a frequency of words within the first embeddings; validating thefirst namebasedoncomparing the first name to a set of merchant names; or determining the set of merchant names based on additional data. 15. The computer-readable, non-transitory medium of claim 14, wherein: determining the first name for the first cluster includes determining the set ofmerchant names based on the additional data; and the instructions further cause the one or more processors to modify the first name based on a similarity comparison between the first name and a merchant name of the set of merchant names. 5 10 15 20 25 30 35 40 45 50 55 15 EP 4 647 998 A1 16 EP 4 647 998 A1 17 EP 4 647 998 A1 18 EP 4 647 998 A1 19 EP 4 647 998 A1 20 EP 4 647 998 A1 21 EP 4 647 998 A1 22 EP 4 647 998 A1 23 EP 4 647 998 A1 24 EP 4 647 998 A1 25 EP 4 647 998 A1 5 10 15 20 25 30 35 40 45 50 55 26 EP 4 647 998 A1 5 10 15 20 25 30 35 40 45 50 55 27 EP 4 647 998 A1 REFERENCES CITED IN THE DESCRIPTION This list of references cited by the applicant is for the reader’s convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard. Patent documents cited in the description • US 63645404

[0001] 摘要 一種系統，包括一個或多個處理器，以及包含指令的計算機可讀非暫態介質。所 述指令使得所述一個或多個處理器中的至少一個執行以下操作：獲取包含多個商 戶的商戶數據；獲取使用大語言模型提取的一組詞嵌入；通過以所述商戶數據為 輸入執行機器學習模型來精煉所述一組詞嵌入，以獲得一組商戶嵌入；在所述一 組商戶嵌入中確定第一商戶嵌入的第一聚類和第二商戶嵌入的第二聚類；基於所 述第一嵌入為所述第一聚類確定第一名稱，並基於所述第二嵌入為所述第二聚類 確定第二名稱；以及基於所述第一名稱與所述第二名稱的相似性，合併所述第一 聚類與所述第二聚類，以獲得合併聚類。

Claims

1. A system comprising: one or more processors; and a computer-readable, non-transitory medium including instructions which, when executed by the one or more processors, cause at least one of the one or more processors to: obtain merchant data including a plurality of merchants; obtain a set of word embeddings extracted using a large language model; refine the set of word embeddings by executing a machine-learning model using as input the merchant data to obtain a set of merchant embeddings; determine a first cluster of first merchant embeddings and a second cluster of second merchant embeddings within the set of merchant embeddings; determine a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the second embeddings; and merge the first cluster and the second cluster based on a similarity of the first name and the second name to obtain a merged cluster.

2. The system of claim 1, wherein refining the set of word embeddings includes: generating, by the machine-learning model, a predicted category for each word embedding of the set of word embeddings; and refining the set of word embeddings based on a comparison of the predicted category for each word embedding and a corresponding category label in the merchant data.

3. The system of claim 1 or 2, wherein refining the set of word embeddings includes: determining a distance between a first merchant embedding and a second merchant embedding; and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding.

4. The system of any of the preceding claims, wherein determining the first name for the first cluster includes: determining the first name for the first cluster based on a frequency of words within the first embeddings; validating the first name based on comparing the first name to a set of merchant names; or determining the set of merchant names based on additional data.

5. The system of claim 4, wherein: determining the first name for the first cluster includes determining the set of merchant names based on the additional data; and the instructions further cause the one or more processors to modify the first name based on a similarity comparison between the first name and a merchant name and the set of merchant names.

6. A method comprising: obtaining merchant data including a plurality of merchants; obtaining a set of word embeddings extracted using a large language model; refining the set of word embeddings by executing a machine-learning model using as input the merchant data to obtain a set of merchant embeddings; determining a first cluster of first merchant embeddings and a second cluster of second merchant embeddings within the set of merchant embeddings; determining a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the second embeddings; and merging the first cluster and the second cluster based on a similarity of the first name and the second name to obtain a merged cluster.

7. The method of claim 6, wherein refining the set of word embeddings includes: generating, by the machine-learning model, a predicted category for each word embedding of the set of word embeddings; and refining the set of word embeddings based on a comparison of the predicted category for each word embedding and a corresponding category label in the merchant data.

8. The method of claim 6 or 7, wherein refining the set of word embeddings includes: determining a distance between a first merchant embedding and a second merchant embedding; and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding.

9. The method of any of claims 6 to 8, wherein determining the first name for the first cluster includes: determining the first name for the first cluster based on a frequency of words within the first embeddings; validating the first name based on comparing the first name to a set of merchant names; or determining the set of merchant names based on additional data.

10. The method of claim 9, wherein: determining the first name for the first cluster includes determining the set of merchant names based on the additional data; and the method further comprises modifying the first name based on a similarity comparison between the first name and a merchant name of the set of merchant names.

11. A computer-readable, non-transitory medium including instructions which, when executed by one or more processors, cause at least one of the one or more processors to: obtain merchant data including a plurality of merchants; obtain a set of word embeddings extracted using a large language model; refine the set of word embeddings by executing a machine-learning model using as input the merchant data to obtain a set of merchant embeddings; determine a first cluster of first merchant embeddings and a second cluster of second merchant embeddings within the set of merchant embeddings; determine a first name for the first cluster based on the first embeddings and a second name for the second cluster based on the second embeddings; and merge the first cluster and the second cluster based on a similarity of the first name and the second name to obtain a merged cluster.

12. The computer-readable, non-transitory medium of claim 11, wherein refining the set of word embeddings includes: generating, by the machine-learning model, a predicted category for each word embedding of the set of word embeddings; and refining the set of word embeddings based on a comparison of the predicted category for each word embedding and a corresponding category label in the merchant data.

13. The computer-readable, non-transitory medium of claim 11 or 12, wherein refining the set of word embeddings includes: determining a distance between a first merchant embedding and a second merchant embedding; and applying a loss function to reduce a difference between the determined distance and a labeled distance between the first merchant embedding and the second merchant embedding.

14. The computer-readable, non-transitory medium of any of claims 11 to 13, wherein determining the first name for the first cluster includes: determining the first name for the first cluster based on a frequency of words within the first embeddings; validating the first name based on comparing the first name to a set of merchant names; or determining the set of merchant names based on additional data.

15. The computer-readable, non-transitory medium of claim 14, wherein: determining the first name for the first cluster includes determining the set of merchant names based on the additional data; and the instructions further cause the one or more processors to modify the first name based on a similarity comparison between the first name and a merchant name of the set of merchant names.

Images