# US 2025/0148482 A1 · Systems and methods for dynamically updating models using machine learning

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

- Number: US 2025/0148482 A1
- Document ID: US20250148482A1
- Status: Published application
- Kind: A1
- Application number: 18/501,855
- Filed: 2023-11-03
- Published: 2025-05-08
- Earliest claimed: 2023-11-03
- Assignees: Mastercard International Incorporation
- Applicants: Mastercard International Incorporation
- Inventors: Joshua A. Allbright; Christopher J. Merz
- Web page: https://patentgrep.com/patent/US20250148482A1

## CPC

- G06Q 30/0185: {Product, service or business identity fraud}
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N 20/20: Ensemble learning
- 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

A computing system for detecting patterns in data is provided. The computing system includes a model engine configured to receive an initial dataset, and segment the initial dataset into a plurality of subsets. The model engine is further configured to assign a weight to each subset based at least in part on an age of the subset, train a machine learning model on each subset separately in accordance with the assigned weighting for that subset. The model engine is further configured to receive a candidate dataset, analyze the candidate dataset using the trained machine learning model, and assign a score to the candidate dataset based on the analysis. The computing system further includes a rules engine configured to receive the candidate dataset and the corresponding score from the model engine, and generate and output, based at least in part on the score, a decision regarding the candidate dataset.

## Claims

### Claim 1

A computing system for detecting patterns in data transmitted over a network, said computing system comprising:
a model engine configured to:
receive, from a database, an initial dataset including historical data for a first time period;
segment the initial dataset into a plurality of subsets, each subset associated with a second time period that is smaller than the first time period;
assign a weight to each of the plurality of subsets, each weight assigned based at least in part on an age of the associated subset;
train a machine learning model on each subset of the plurality of subsets separately, wherein for each subset, the machine learning model is trained on that subset in accordance with the assigned weighting for that subset;
receive, from a computing device, a candidate dataset;
analyze the candidate dataset using the trained machine learning model; and
assign a score to the candidate dataset based on the analysis; and
a rules engine communicatively coupled to the model engine and configured to:
receive the candidate dataset and the corresponding score from the model engine; and
generate and output, based at least in part on the score, a decision regarding the candidate dataset.

### Claim 2 (depends on 1)

The computing system of claim 1, wherein the machine learning model is a gradient-boosted decision tree model.

### Claim 3 (depends on 1)

The computing system of claim 1, wherein to train the machine learning model, the model engine is configured to update probabilities associated with each leaf node of the machine learning model when training on each subset.

### Claim 4 (depends on 1)

The computing system of claim 1, wherein to assign a weight, the model engine is configured to assign a lower weight to more recent subsets, such that the machine learning model trains less on more recent subsets.

### Claim 5 (depends on 1)

The computing system of claim 1, wherein the first time period is one year, and wherein each second time period is one month.

### Claim 6 (depends on 1)

The computing system of claim 1, wherein the machine learning model is a fraud scoring model.

### Claim 7

A computing system for detecting and preventing fraudulent network events in a payment card network, said computing system comprising:
a fraud model engine configured to:
receive, from a database, an initial dataset including historical transaction data for a first time period;
segment the initial dataset into a plurality of subsets, each subset associated with a second time period that is smaller than the first time period;
assign a weight to each of the plurality of subsets, each weight assigned based at least in part on an age of the associated subset;
train a fraud scoring model on each subset of the plurality of subsets separately, wherein for each subset, the fraud scoring model is trained on that subset in accordance with the assigned weighting for that subset;
receive, from a merchant computing device, a payment card transaction request;
analyze the payment card transaction request using the trained fraud scoring model; and
assign a score to the payment card transaction request based on the analysis; and
a fraud rules engine communicatively coupled to the fraud model engine and configured to:
receive the payment card transaction request and the corresponding score from the fraud model engine; and
generate and output, based at least in part on the score, a decision whether to approve or decline a transaction associated with the payment card transaction request.

### Claim 8 (depends on 7)

The computing system of claim 7, wherein the fraud scoring model is a machine learning model.

### Claim 9 (depends on 7)

The computing system of claim 7, wherein the fraud scoring model is a gradient-boosted decision tree model.

### Claim 10 (depends on 7)

The computing system of claim 7, wherein to train the fraud scoring model, the fraud model engine is configured to update probabilities associated with each leaf node of the fraud scoring model when training on each subset.

### Claim 11 (depends on 7)

The computing system of claim 7, wherein to assign a weight, the fraud model engine is configured to assign a lower weight to more recent subsets, such that the fraud scoring model trains less on more recent subsets.

### Claim 12 (depends on 7)

The computing system of claim 7, wherein the first time period is one year, and wherein each second time period is one month.

### Claim 13 (depends on 7)

The computing system of claim 7, wherein the fraud scoring model is a supervised machine learning model.

### Claim 14

A computer-implemented method for detecting and preventing fraudulent network events in a payment card network, said method comprising:
receiving, at a fraud model engine, from a database, an initial dataset including historical transaction data for a first time period;
segmenting the initial dataset into a plurality of subsets, each subset associated with a second time period that is smaller than the first time period;
assigning a weight to each of the plurality of subsets, each weight assigned based at least in part on an age of the associated subset;
training a fraud scoring model on each subset of the plurality of subsets separately, wherein for each subset, the fraud scoring model is trained on that subset in accordance with the assigned weighting for that subset;
receiving, from a merchant computing device, a payment card transaction request;
analyzing the payment card transaction request using the trained fraud scoring model;
assigning a score to the payment card transaction request based on the analysis;
receiving, at a fraud rules engine communicatively coupled to the fraud model engine, the payment card transaction request and the corresponding score from the fraud model engine; and
generating and outputting, based at least in part on the score, a decision whether to approve or decline a transaction associated with the payment card transaction request.

### Claim 15 (depends on 14)

The method of claim 14, wherein the fraud scoring model is a machine learning model.

### Claim 16 (depends on 14)

The method of claim 14, wherein the fraud scoring model is a gradient-boosted decision tree model.

### Claim 17 (depends on 14)

The method of claim 14, wherein training the fraud scoring model comprises updating probabilities associated with each leaf node of the fraud scoring model when training on each subset.

### Claim 18 (depends on 14)

The method of claim 14, wherein assigning a weight comprises assigning a lower weight to more recent subsets, such that the fraud scoring model trains less on more recent subsets.

### Claim 19 (depends on 14)

The method of claim 14, wherein the first time period is one year, and wherein each second time period is one month.

### Claim 20 (depends on 14)

The method of claim 14, wherein the fraud scoring model is a supervised machine learning model.
