Digital Direct Deposit Predictor Machine Learning Model

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional transaction management systems face inaccuracies and inefficiencies in determining the risk involved in providing digital direct deposit advances due to reliance on rules-based or heuristic methods that fail to incorporate changing variables and are inflexible.

Innovation Solution

The implementation of a digital direct deposit predictor machine-learning model that generates digital direct deposit likelihoods by analyzing features of network transactions, incorporating additional user account data, and using a risk analysis assembler to process transactions intelligently.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If rules-based or heuristic methods are used to determine digital direct deposit advance amounts, then the system is simple to implement, but the accuracy of risk prediction deteriorates

Engineering Contradiction:

Improveease of implementationVSAvoidaccuracy of risk prediction

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent replaces rules-based or heuristic methods (mechanical system) with a machine-learning model that uses automated data processing and algorithmic predictions. The machine-learning model processes multiple data sources including transaction history, account information, and external data to generate accurate risk predictions without relying on predefined rules

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the parameters used for risk assessment by incorporating multiple data sources and features beyond traditional historical deposit transaction data. The machine-learning model evaluates numerous variables including user behavior patterns, transaction characteristics, and external factors to dynamically adjust risk predictions

Inventive Principle:

Principle #35Parameter changes

2Loss of information

If multiple user interfaces are used to gather data for digital direct deposit advances, then comprehensive information can be collected, but computational resource efficiency deteriorates

Engineering Contradiction:

Improvecompleteness of informationVSAvoidcomputational resource efficiency

Core Design Contradiction:

Loss of informationVSProductivity

Solution Approach 1:

The patent merges multiple data gathering functions into a single streamlined user interface. The machine-learning model automatically accesses and integrates data from multiple sources including transaction history, account information, and external data sources without requiring users to navigate multiple interfaces or manually input information

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The system performs self-service by automatically gathering, processing, and analyzing data from multiple sources without requiring user intervention. The machine-learning model autonomously accesses necessary information and generates risk predictions, eliminating the need for users to provide detailed information through multiple interfaces

Inventive Principle:

Principle #25Self-service

3Device complexity

If only historical deposit transaction data is used for risk assessment, then data collection is simple, but the flexibility and robustness of risk analysis deteriorates

Engineering Contradiction:

Improvesimplicity of data collectionVSAvoidflexibility of risk analysis

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent implements a multi-functional data collection system that gathers information from diverse sources including transaction history, account information, user behavior data, and external data sources. The machine-learning model processes this diverse data to perform multiple functions including risk assessment, fraud detection, and creditworthiness evaluation

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system transitions from static historical data to dynamic multi-source data collection. The machine-learning model continuously processes changing variables and adapts to new patterns, allowing the risk analysis system to remain flexible and robust in response to evolving user behaviors and transaction patterns

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS20250139631A1Utilizing a digital direct deposit predictor machine-learning model to determine risk for a network transaction comprising a digital direct deposit advance

Publication Date: 2025.05.01 CHIME FINANCIAL INC

AI summary

The present disclosure relates to systems, non-transitory computer-readable media, and methods for utilizing a digital direct deposit predictor machine-learning model to generate a digital direct deposit likelihood and processing a network transaction based on the digital direct deposit likelihood. In particular, in one or more embodiments, the disclosed systems generate a risk classification based on the digital direct deposit likelihood and utilize the risk classification to process the network transaction. Moreover, in one or more embodiments, the disclosed systems generate a digital direct deposit amount based on the digital direct deposit likelihood and/or the risk classification and process the network transaction accordingly. Moreover, the disclosed systems can display information related to network transactions in a digital direct deposit interface.