Prediction Model for Transaction Exception Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer-automated systems fail to accurately and intelligently classify and narrate transaction exceptions, leading to inefficiencies in transaction processing, increased costs, and reduced customer satisfaction.
Innovation Solution
A neural network or prediction model is used to identify transactions as exceptions, predict classifications, generate rules, and create decision graphs to assign classifications and provide narrations, leveraging resolved exceptions information to process transaction information and improve classification accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional computer-automated systems are used for transaction matching, then basic transaction processing can be performed, but classification and narration of exceptions cannot be done intelligently and accurately
Solution Approach 1:
The patent replaces traditional rule-based mechanical classification systems with a neural network-based intelligent system. The neural network learns patterns from historical resolved exceptions data and automatically classifies new exceptions with high accuracy, eliminating the need for manual rule configuration and achieving both precision and adaptability simultaneously.
Solution Approach 2:
The system transforms the classification approach by changing from static rule parameters to dynamic learned parameters. The neural network continuously adapts its internal parameters (weights and biases) based on training data, enabling accurate classification across diverse exception types without requiring explicit rule definitions for each scenario.
2Measurement precision
If more resolved exceptions information is used as input to the prediction model, then classification accuracy improves, but data processing complexity increases
Solution Approach 1:
The system performs preliminary action by pre-processing and structuring the resolved exceptions data before feeding it to the neural network. Historical data is cleaned, normalized, and organized into appropriate feature formats in advance, which simplifies the ongoing processing complexity while maintaining high classification accuracy through comprehensive training data.
Solution Approach 2:
The neural network creates an internal copy or representation of the complex relationships in the training data through its learned weight structures. This internal model captures the essential patterns from large volumes of resolved exceptions without requiring the system to repeatedly process the raw complex data, thus maintaining accuracy while reducing processing complexity during inference.
Data Source
AI summary
In certain embodiments, resolved exceptions information regarding resolved exceptions may be obtained. The resolved exceptions information may indicate the resolved exceptions and, for each resolved exception of the resolved exceptions, a set of attributes of a transaction for which the resolved exception was triggered. The resolved exceptions information may be provided as input to a prediction model to obtain multiple decision trees via the prediction model. Each decision tree of the multiple decision trees may comprise nodes and conditional branches, each node of the nodes of the decision tree indicating a probability of a dividend-related classification for a transaction that corresponds to the node. A decision tree may be obtained from the multiple decision trees. Unresolved exception information regarding unresolved exceptions may be processed based on the decision tree to determine which of the decision tree's nodes respectively correspond to transactions for which the unresolved exceptions were triggered.