Machine Learning Data Source Evaluation for Credit Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing credit evaluation systems face challenges in assessing the creditworthiness of applicants with insufficient domestic credit history or low credit scores, as they rely primarily on credit bureau data, which may not provide a comprehensive view of an applicant's financial behavior.
Innovation Solution
A decisioning system that utilizes machine learning techniques to evaluate and recommend alternative data sources, allowing applicants to select additional data sources such as bank account assets, academic data, bill payment histories, and rental payment histories, to provide a more holistic assessment of creditworthiness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If credit evaluation systems rely primarily on credit bureau data, then the evaluation process is simple and fast, but the assessment of creditworthiness is incomplete for applicants with insufficient domestic credit history
Solution Approach 1:
The patent segments the data sources into multiple categories including domestic credit bureau data, international credit bureau data, and alternative data sources. This segmentation allows the system to evaluate different types of data separately and combine them strategically, improving creditworthiness assessment accuracy without overwhelming complexity by handling each data type through dedicated processing modules.
Solution Approach 2:
The machine learning model serves multiple functions: it evaluates the effectiveness of different data sources, predicts creditworthiness, and optimizes data source selection. This multi-functionality improves assessment accuracy while managing system complexity by consolidating multiple evaluation tasks into a single unified model rather than requiring separate specialized systems for each function.
2Loss of information
If alternative data sources are added to evaluate creditworthiness, then the comprehensive view of applicant financial behavior improves, but the system complexity and data processing requirements increase
Solution Approach 1:
The system implements feedback mechanisms where the machine learning model continuously learns from the effectiveness of different data sources in predicting creditworthiness. This feedback loop allows the system to automatically optimize which data sources to use for specific applicant profiles, reducing manual configuration complexity while maintaining comprehensive financial behavior assessment through multiple data types.
Solution Approach 2:
The system performs preliminary evaluation of alternative data sources using machine learning models before actually using them in credit decisions. By pre-evaluating and ranking data sources based on their effectiveness for different applicant profiles, the system reduces the complexity of real-time data source selection and maintains comprehensive information gathering through strategic pre-filtering of relevant data types.
3Measurement precision
If multiple alternative data sources are evaluated, then the credit decision accuracy improves, but the time and resources required for data collection and processing increase
Solution Approach 1:
The system applies partial action by selectively evaluating and using only the most relevant alternative data sources for each applicant profile rather than comprehensively collecting all possible data types. The machine learning model identifies and processes only the necessary data sources needed to achieve accurate credit decisions, reducing processing time and resource consumption while maintaining high measurement precision through targeted data selection.
Data Source
AI summary
In some implementations, a decisioning system may use a machine learning model to generate a recommended set of alternative data sources for providing information related to behavioral attributes of a user of a client device. The decisioning system may present, to the client device, an interface that indicates that the recommended set of alternative data sources. The decisioning system may receive, from the client device, a selection of one or more alternative data sources for providing information related to the behavioral attributes of the user. The decisioning system may obtain information related to the behavioral attributes of the user from the selected alternative data sources and may generate a decision associated with the application based on the information obtained from the one or more alternative data sources.