# US 2025/0094971 A1 · Artificial intelligence-based methods and systems for generating optimal embeddings for identifying similar entities

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0094971 A1
- Document ID: US20250094971A1
- Status: Published application
- Kind: A1
- Application number: 18/471,060
- Filed: 2023-09-20
- Published: 2025-03-20
- Earliest claimed: 2023-09-20
- Assignees: Mastercard International Incorporated
- Applicants: Mastercard International Incorporated
- Inventors: Gaurav Oberoi; Pranav Poduval; FNU Karamjit Singh
- Web page: https://patentgrep.com/patent/US20250094971A1

## CPC

- G06Q 20/389: {Keeping log of transactions for guaranteeing non-repudiation of a transaction}
- G06N 3/045: Combinations of networks
- G06Q 20/325: {using wireless networks}
- G06Q 20/42: Confirmation, e.g. check or permission by the legal debtor of payment
- G06Q 40/06: Asset management; Financial planning or analysis
- G06Q 40/12: Accounting
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Methods and server systems for performing similarity-search are described herein. Method performed by server system includes accessing entity-related dataset and using that generating set of entity-specific features including subset of numerical features and subset of categorical features corresponding to each of plurality of entities. Method includes determining label corresponding to each of plurality of entities based on subset of categorical features. Method includes generating via first model, first set of entity-specific embeddings corresponding to each of plurality of entities based on corresponding subset of numerical features and corresponding label. Method includes generating via second model, second set of entity-specific embeddings corresponding to each of plurality of entities based on corresponding subset of categorical features and corresponding first set of entity-specific embeddings, further generating set of optimal embeddings corresponding to each of plurality of entities based on concatenating corresponding first set of entity-specific embeddings and corresponding second set of entity-specific embeddings.

## Claims

### Claim 1

A computer-implemented method, comprising:
accessing, by a server system, an entity-related dataset from a database associated with the server system, the entity-related dataset comprising information related to a plurality of entities;
generating, by the server system, a set of entity-specific features corresponding to each of the plurality of entities based, at least in part, on the entity-related dataset, the set of entity-specific features comprising a subset of numerical features and a subset of categorical features;
determining, by the server system, a label corresponding to each of the plurality of entities based, at least in part, on the subset of categorical features;
generating, by the server system via a first machine learning model, a first set of entity-specific embeddings corresponding to each of the plurality of entities based, at least in part, on the corresponding subset of numerical features and the corresponding label;
generating, by the server system via a second machine learning model, a second set of entity-specific embeddings corresponding to each of the plurality of entities based, at least in part, on the corresponding subset of categorical features and the corresponding first set of entity-specific embeddings; and
generating, by the server system, a set of optimal embeddings corresponding to each of the plurality of entities based, at least in part, on concatenating the corresponding first set of entity-specific embeddings and the corresponding second set of entity-specific embeddings.

### Claim 2 (depends on 1)

The computer-implemented method as claimed in claim 1, wherein generating the first set of entity-specific embeddings, further comprises:
extracting, by the server system, a set of important features from the subset of numerical features based, at least in part, on a task and a set of pre-defined rules;
computing, by the server system, anew set of weights for each of the plurality of entities based, at least in part, on adjusting weights associated with the set of important features; and
generating, by the server system via the first machine learning model, the first set of entity-specific embeddings based, at least in part, on the new set of weights calculated for each of the plurality of entities and the corresponding label.

### Claim 3 (depends on 1)

The computer-implemented method as claimed in claim 1, wherein generating the second set of entity-specific embeddings, further comprises:
generating, by the server system, a set of intermediate entity-specific embeddings corresponding to each of the plurality of entities based, at least in part, on the corresponding subset of categorical features; and
determining, by the server system via a self-learning attention process, the second set of entity-specific embeddings corresponding to each of the plurality of entities by updating the corresponding set of intermediate entity-specific embeddings based, at least in part, on the corresponding first set of entity-specific embeddings and the subset of categorical features.

### Claim 4 (depends on 1)

The computer-implemented method as claimed in claim 1, further comprising:
generating, by the server system, a homogeneous entity graph based, at least in part, on the entity-related dataset and the set of entity-specific features for each of the plurality of entities, the homogeneous entity graph comprising a plurality of nodes, each of the plurality of nodes corresponding to each of the plurality of entities, wherein each of the plurality of nodes comprises the corresponding set of entity-specific features.

### Claim 5 (depends on 1)

The computer-implemented method as claimed in claim 1, further comprising:
generating, by the server system, an entity-specific graph based, at least in part, on the set of optimal embeddings corresponding to each of the plurality of entities.

### Claim 6 (depends on 5)

The computer-implemented method as claimed in claim 5, further comprising:
determining, by the server system, a similarity range for an entity from the plurality of entities based, at least in part, on a task; and
generating, by the server system, a peer set for the entity based, at least in part, on the entity-specific graph, and the similarity range, the peer set indicating a set of entities from the entity-specific graph within the similarity range from the entity.

### Claim 7 (depends on 1)

The computer-implemented method as claimed in claim 1, wherein the first machine learning model and the second machine learning model are graph neural network (GNN) based machine learning models.

### Claim 8 (depends on 1)

The computer-implemented method as claimed in claim 1, wherein the entity-related dataset is a historical transaction dataset, the historical transaction dataset comprising information related to a plurality of historical payment transactions performed between a plurality of cardholders and a plurality of merchants.

### Claim 9 (depends on 1)

The computer-implemented method as claimed in claim 1, wherein the plurality of entities is one of a plurality of merchants, a plurality of cardholders, a plurality of issuers, and a plurality of acquirers.

### Claim 10 (depends on 1)

The computer-implemented method as claimed in claim 1, wherein the server system is a payment server associated with a payment network.

### Claim 11

A server system, comprising:
a memory configured to store instructions;
a communication interface; and
a processor in communication with the memory and the communication interface, the processor configured to execute the instructions stored in the memory and thereby cause the server system to perform at least in part to:
access an entity-related dataset from a database associated with the server system, the entity-related dataset comprising information related to a plurality of entities;
generate a set of entity-specific features corresponding to each of the plurality of entities based, at least in part, on the entity-related dataset, the set of entity-specific features comprising a subset of numerical features and a subset of categorical features;
determine a label corresponding to each of the plurality of entities based, at least in part, on the subset of categorical features;
generate via a first machine learning model, a first set of entity-specific embeddings corresponding to each of the plurality of entities based, at least in part, on the corresponding subset of numerical features and the corresponding label;
generate via a second machine learning model, a second set of entity-specific embeddings corresponding to each of the plurality of entities based, at least in part, on the corresponding subset of categorical features and the corresponding first set of entity-specific embeddings; and
generate a set of optimal embeddings corresponding to each of the plurality of entities based, at least in part, on concatenating the corresponding first set of entity-specific embeddings and the corresponding second set of entity-specific embeddings.

### Claim 12 (depends on 11)

The server system as claimed in claim 11, wherein for generating the first set of entity-specific embeddings, the server system is further caused, at least in part, to:
extract a set of important features from the subset of numerical features based, at least in part, on a task;
compute a new set of weights for each of the plurality of entities based, at least in part, on adjusting weights associated with the set of important features; and
generate via the first machine learning model, the first set of entity-specific embeddings based, at least in part, on the corresponding new set of weights calculated for each of the plurality of entities and the corresponding label.

### Claim 13 (depends on 11)

The server system as claimed in claim 11, wherein for generating the second set of entity-specific embeddings, the server system is further caused, at least in part, to:
generate a set of intermediate entity-specific embeddings corresponding to each of the plurality of entities based, at least in part, on the corresponding subset of categorical features; and
determine via a self-learning attention process, the second set of entity-specific embeddings corresponding to each of the plurality of entities by updating the corresponding set of intermediate entity-specific embeddings based, at least in part, on the corresponding first set of entity-specific embeddings and the subset of categorical features.

### Claim 14 (depends on 11)

The server system as claimed in claim 11, wherein the server system is further caused, at least in part, to:
generate a homogeneous entity graph based, at least in part, on the entity-related dataset and the set of entity-specific features for each of the plurality of entities, the homogeneous entity graph comprising a plurality of nodes, each of the plurality of nodes corresponding to each of the plurality of entities, wherein each of the plurality of nodes comprises the corresponding set of entity-specific features.

### Claim 15 (depends on 11)

The server system as claimed in claim 11, wherein the server system is further caused, at least in part, to:
generate an entity-specific graph based, at least in part, on the set of optimal embeddings corresponding to each of the plurality of entities.
determine a similarity range for an entity from the plurality of entities based, at least in part, on a task; and
generate a peer set for the entity based, at least in part, on the entity-specific graph, and the similarity range, the peer set indicating a set of entities from the entity-specific graph within the similarity range from the entity.

### Claim 16 (depends on 11)

The server system as claimed in claim 11, wherein the first machine learning model and the second machine learning model are graph neural network (GNN) based machine learning models.

### Claim 17 (depends on 11)

The server system as claimed in claim 11, wherein the entity-related dataset is a historical transaction dataset, the historical transaction dataset comprising information related to a plurality of historical payment transactions performed between a plurality of cardholders and a plurality of merchants.

### Claim 18 (depends on 11)

The server system as claimed in claim 11, wherein the plurality of entities is one of a plurality of merchants, a plurality of cardholders, a plurality of issuers, and a plurality of acquirers.

### Claim 19

A non-transitory computer-readable storage medium comprising computer-executable instructions that, when executed by at least a processor of a server system, cause the server system to perform a method comprising:
accessing an entity-related dataset from a database associated with the server system, the entity-related dataset comprising information related to a plurality of entities;
generating a set of entity-specific features corresponding to each of the plurality of entities based, at least in part, on the entity-related dataset, the set of entity-specific features comprising a subset of numerical features and a subset of categorical features;
determining a label corresponding to each of the plurality of entities based, at least in part, on the subset of categorical features;
generating via a first machine learning model, a first set of entity-specific embeddings corresponding to each of the plurality of entities based, at least in part, on the corresponding subset of numerical features and the corresponding label;
generating via a second machine learning model, a second set of entity-specific embeddings corresponding to each of the plurality of entities based, at least in part, on the corresponding subset of categorical features and the corresponding first set of entity-specific embeddings; and
generating a set of optimal embeddings corresponding to each of the plurality of entities based, at least in part, on concatenating the corresponding first set of entity-specific embeddings and the corresponding second set of entity-specific embeddings.

### Claim 20 (depends on 19)

The non-transitory computer-readable storage medium as claimed in claim 19, wherein for generating the first set of entity-specific embeddings, the method further comprises:
extracting a set of important features from the subset of numerical features based, at least in part, on a task;
compute a new set of weights for each of the plurality of entities based, at least in part, on adjusting weights associated with the set of important features; and
generating via the first machine learning model, the first set of entity-specific embeddings based, at least in part, on the corresponding new set of weights calculated for each of the plurality of entities and the corresponding label.