Credit Score Normalization Across Multiple Aggregators
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current credit scoring models are limited by their reliance on data from a single aggregator and credit reporting agency, leading to inflexible and less accurate credit assessments, as they cannot effectively process consumer permissioned financial data from multiple sources due to differences in data structures, which can result in less informed lending decisions.
Innovation Solution
A method to normalize consumer permissioned financial data and credit file data from various aggregators and credit reporting agencies, allowing for the use of a common classification system to establish attribute values in credit scoring models, regardless of the data structure, enabling the generation of updated credit scores that are more accurate and flexible.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If credit scoring models use data from a single aggregator and credit reporting agency, then the data processing is simple and straightforward, but the credit assessment becomes inflexible and less accurate
Solution Approach 1:
The patent creates a universal credit scoring system that can process financial data from multiple aggregators and credit reporting agencies through a common platform. The system uses standardized data fields and normalization processes to handle diverse data sources, allowing the same credit scoring model to work with data from any aggregator without requiring separate processing pipelines for each source.
Solution Approach 2:
The system transforms data from different aggregators by normalizing their unique data structures into a common format. This involves mapping various data fields from different sources to standardized parameters, allowing the credit scoring model to receive consistent input regardless of the original data source's structure or naming conventions.
2Adaptability or versatility
If the system processes consumer permissioned financial data from multiple aggregators, then the credit score accuracy and flexibility improve, but the data structure differences create processing challenges
Solution Approach 1:
The patent implements homogeneity by converting heterogeneous data from multiple aggregators into a uniform format. The system defines common data fields that all aggregators must provide, and automatically transforms their diverse data structures into this standardized format, creating homogeneous input for the credit scoring model while maintaining the ability to accept data from any source.
3Ease of manufacture
If creditors rely on aggregators with existing processes, then the integration is easier, but the system becomes limited to a single aggregator and loses flexibility
Solution Approach 1:
The patent introduces a intermediary normalization layer between the credit scoring model and multiple aggregators. This intermediary component translates data from various aggregator formats into a universal format that the scoring model can process, allowing creditors to switch between aggregators or use multiple aggregators simultaneously without redesigning the entire system architecture.
Data Source
AI summary
A method employs consumer permissioned financial data to generate an updated credit score following a consumer authorization to access to at least one demand deposit account of the consumer for use in generating the credit score for the consumer. The consumer permissioned financial data may be sourced from any one of a plurality of aggregators where the aggregators employ different data structures for the financial data. The financial data is processed to automatically normalize the data received from an aggregator by applying a common classification system across the plurality of aggregators to classify a first set of data fields; and automatically, regardless of whether an updated credit score is advantageous for the consumer and independent of any further consumer permission, generating the updated credit score using a combination of credit file data for the consumer and the financial data received from the aggregator.