Server systems and methods for merchant data cleansing in payment network

Patent metadata

Field	Value
Publication number	US-11989165-B2
Application number	US-202217879680-A
Country	US
Kind code	B2
Filing date	Aug 2, 2022
Priority date	Aug 4, 2021
Publication date	May 21, 2024
Grant date	May 21, 2024

Abstract

Embodiments provide methods and systems for merchant data cleansing in payment network. Method performed by server system includes accessing electronic payment transaction records from transaction database. Each electronic payment transaction record includes merchant data fields. Method includes determining set of electronic payment transaction records with ambiguous merchant data fields having matching probability scores less than predetermined threshold value computed by probabilistic matching model and identifying at least one issue for non-matching of each of set of electronic payment transaction records. Method includes determining data model based on at least one issue of each of set of electronic payment transaction records. Data model is one of: phone-to-city model, payment aggregator model, and merchant name normalization model. Method includes updating set of electronic payment transaction records with unambiguous merchant data fields corresponding to ambiguous merchant data fields by applying data model to each of set of electronic payment transaction records.

First claim

The invention claimed is:

1. A computer-implemented method comprising:
accessing, by a server system, a plurality of electronic payment transaction records associated with a plurality of merchants from a transaction database, each of the plurality of electronic payment transaction records comprising merchant data fields associated with a merchant of the plurality of merchants;
determining, by the server system, a set of electronic payment transaction records with ambiguous merchant data fields, each of the set of electronic payment transaction records from the plurality of electronic payment transaction records having a matching probability score less than a predetermined threshold value, wherein the matching probability score is computed by a probabilistic matching model;
identifying, by the server system, at least one issue for non-matching of each of the set of electronic payment transaction records;
determining, by the server system, at least one data model based, at least in part, on the at least one issue of each of the set of electronic payment transaction records, wherein the at least one data model is one of: phone-to-city model, payment aggregator model, and merchant name normalization model;
updating, by the server system, the set of electronic payment transaction records having ambiguous merchant data fields with corresponding unambiguous merchant data fields by applying the at least one data model to each of the set of electronic payment transaction records;
applying, by the server system, the merchant name normalization model over third electronic payment transaction records to determine aggregated merchant names, wherein merchant name fields of the third electronic payment transaction records are populated with ambiguous merchant names, and wherein the merchant name normalization model is based on a transformer neural network model with character level encoding; and
updating, by the server system, the merchant name fields of the third electronic payment transaction records based on the application of the merchant name normalization model over the third electronic payment transaction records.

2. The computer-implemented method as claimed in claim 1 , further comprising:
pre-processing, by the server system, the plurality of electronic payment transaction records based, at least in part, on a predefined ruleset.

3. The computer-implemented method as claimed in claim 1 , wherein the probabilistic matching model is a logistic regression model.

4. The computer-implemented method as claimed in claim 1 , wherein the at least one issue is one of: phone number in a city name field, payment aggregators sending the ambiguous merchant data fields, and ambiguous merchant name in merchant name field.

5. The computer-implemented method as claimed in claim 1 , further comprising:
applying, by the server system, the phone-to-city model over first electronic payment transaction records in which city name fields are populated with phone numbers of merchants associated with the first electronic payment transaction records, the phone-to-city model configured to predict city names with prediction scores against the first electronic payment transaction records; and
updating, by the server system, the city name fields of the first electronic payment transaction records with the predicted city names having prediction scores greater than a threshold value.

6. The computer-implemented method as claimed in claim 1 , further comprising:
applying, by the server system, the payment aggregator model over second electronic payment transaction records in which at least one merchant data field of each second electronic payment transaction record is populated with ambiguous data by payment aggregators, wherein the payment aggregator model comprises Long Short Term Memory (LSTM) neural network with character level encoding; and
updating, by the server system, the at least merchant data field of each second payment transaction records based on the application of the payment aggregator model over the second electronic payment transaction records.

7. The computer-implemented method as claimed in claim 1 , further comprising:
storing, by the server system, the updated set of electronic payment transaction records into the clean merchant database.

8. The computer-implemented method as claimed in claim 1 , wherein the merchant data fields are at least one or more of:
merchant name,
acquirer merchant identifier,
merchant address,
merchant city,
merchant zip code,
merchant state code, and
merchant country.

9. A server system comprising:
a communication interface; and
a processor coupled to the communication interface, the processor configured to
access a plurality of electronic payment transaction records associated with a plurality of merchants from a transaction database, each of the plurality of electronic payment transaction records comprising merchant data fields associated with a merchant of the plurality of merchants;
determine a set of electronic payment transaction records with ambiguous merchant data fields, each of the set of electronic payment transaction records from the plurality of electronic payment transaction records having a matching probability score less than a predetermined threshold value, wherein the matching probability score is computed by a probabilistic matching model;
identify at least one issue for non-matching of each of the set of electronic payment transaction records;
determine at least one data model based, at least in part, on the at least one issue of each of the set of electronic payment transaction records, wherein the at least one data model is one of: phone-to-city model, payment aggregator model, and merchant name normalization model;
update the set of electronic payment transaction records having ambiguous merchant data fields with corresponding unambiguous merchant data fields by applying the at least one data model to each of the set of electronic payment transaction records;
apply the merchant name normalization model over third electronic payment transaction records to determine aggregated merchant names, wherein merchant name fields of the third electronic payment transaction records are populated with ambiguous merchant names, and wherein the merchant name normalization model is based on a transformer neural network model with character level encoding; and
update, by the server system, the merchant name fields of the third electronic payment transaction records based on the application of the merchant name normalization model over the third electronic payment transaction records.

10. The server system of claim 9 , wherein the processor is further configured to:
pre-process the plurality of electronic payment transaction records based, at least in part, on a predefined ruleset.

11. The server system of claim 9 , wherein the probabilistic matching model is a logistic regression model.

12. The server system of claim 9 , wherein the at least one issue is one of: phone number in a city name field, payment aggregators sending the ambiguous merchant data fields, and ambiguous merchant name in merchant name field.

13. The server system of claim 9 , wherein the processor is further configured to:
apply the phone-to-city model over first electronic payment transaction records in which city name fields are populated with phone numbers of merchants associated with the first electronic payment transaction records, the phone-to-city model configured to predict city names with prediction scores against the first electronic payment transaction records; and
update the city name fields of the first electronic payment

Assignees

• Mastercard International Inc

Inventors

• Dubey Shashank
• Dhama Gaurav
• Arora Ankur
• Bishnoi Vikas
• Saraswat Ankur
• Wadhwa Hardik
• Katyal Yatin
• Das Debasmita

Classifications

• G06F16/215 Primary

  Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors · CPC title

• G06F16/2365

  Ensuring data consistency and integrity · CPC title

• G06F16/258

  Data format conversion from or to a database · CPC title