Predictive Model Discriminator for Insurance Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current predictive models for identifying questionable insurance claims rely primarily on determinate data and lack the effectiveness when incorporating indeterminate data, such as narrative text and voice recordings, which can contain crucial information for fraud detection.
Innovation Solution
A computer system that captures and processes both determinate and indeterminate data to train a predictive model, using modules for data storage, indeterminate data capture, and natural language processing to enhance the model's ability to identify potentially fraudulent claims.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If only determinate data is used to train the predictive model, then the model structure remains simple and data processing is straightforward, but the model's effectiveness in identifying fraudulent claims is insufficient
Solution Approach 1:
The patent segments data into two distinct types: determinate data (structured, verifiable facts) and indeterminate data (unstructured narrative text and voice recordings). This segmentation allows the system to process each data type through appropriate methods while integrating them in the predictive model, thereby improving model effectiveness without overwhelming complexity
Solution Approach 2:
The patent introduces an intermediary processing layer that captures indeterminate data from narrative text and voice recordings, converts it into structured formats, and integrates it with determinate data. This intermediary layer bridges the gap between simple data processing and comprehensive fraud detection, enhancing model reliability while managing complexity
2Measurement precision
If indeterminate data such as narrative text and voice recordings is incorporated into the predictive model, then the model's ability to detect fraud patterns is enhanced, but the data processing time and computational resources increase
Solution Approach 1:
The patent applies preliminary action by capturing and preprocessing indeterminate data (narrative text and voice recordings) into structured formats before the predictive model is applied. This preprocessing step converts unstructured data into analyzable formats in advance, reducing the computational burden during actual fraud detection and minimizing processing time delays
Solution Approach 2:
The patent substitutes manual analysis of indeterminate data with automated processing systems that convert narrative text and voice recordings into structured data formats. This mechanical/automated substitution replaces time-consuming manual review processes, significantly reducing data processing time while maintaining or improving detection accuracy
Data Source
AI summary
A computer system includes a data storage module which receives, stores, and provides access to determinate data, raw indeterminate data, and extracted indeterminate data captured by an indeterminate data capture module. The computer system also includes a computer processor, a model training component, and a screening module. The model training component generates the predictive model based upon historical determinate and indeterminate data, and continuously adapts the predictive model with new historical data. The screening module categorizes current claims according to whether they are suitable for predictive analysis by the predictive model. The predictive model is applied to the current claims suitable for predictive analysis to determine a value for each claim indicative of whether the current claim transaction is questionable. The system also includes an output device which outputs the determined value for each claim, and a routing modules which routes claim workflow based on the outputted values.