Bayesian Network for Financial Data Error Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current risk assessment systems for financial derivatives, such as Pre-Settlement Exposure (PSE) Servers, face challenges in accurately identifying the causes of changes in exposure profiles due to the large volume and complexity of data, leading to unexplained anomalies that require manual detection by credit analysts, which is impractical and resource-intensive.

Innovation Solution

A customizable Bayesian belief network and Deep Informative Virtual Assistant (DIVA) system is implemented to diagnose and explain changes in exposure profiles by performing induction and sensitivity analysis, using a Bayesian belief network to model relationships between inputs and outputs, and providing sensitivity analysis and explanation context to identify plausible sources of error.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual detection of data errors by credit analysts is used, then data integrity can be verified, but staff time and resources are excessively consumed

Engineering Contradiction:

Improvedata integrityVSAvoidstaff time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces the mechanical manual review process with an automated computer-based system that uses machine learning models and algorithms to detect data errors, anomalies, and inconsistencies in derivative transaction data, thereby eliminating the need for manual credit analyst intervention while maintaining high detection accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by automatically performing data validation, error detection, and anomaly identification functions that previously required human analysts, allowing the computational system to serve itself in verifying data integrity without external human intervention

Inventive Principle:

Principle #25Self-service

2Quantity of substance

If the volume and complexity of data increases, then more comprehensive risk assessment is achieved, but manual detection becomes impractical

Engineering Contradiction:

Improvedata volumeVSAvoiddetectability

Core Design Contradiction:

Quantity of substanceVSEase of operation

Solution Approach 1:

The patent replaces manual detection mechanisms with automated computational systems capable of processing large volumes of complex derivative transaction data, using machine learning algorithms to identify patterns, anomalies, and errors that would be impossible to detect manually at scale

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the operational parameters from manual review capabilities to automated computational processing, enabling the handling of exponentially larger data volumes and complexities while maintaining or improving detection effectiveness through algorithmic analysis

Inventive Principle:

Principle #35Parameter changes

3Productivity

If automated screening systems are implemented, then productivity increases, but system complexity increases

Engineering Contradiction:

Improveprocessing speedVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent replaces simple manual processes with sophisticated automated systems that use machine learning models, statistical analysis, and computational algorithms to screen derivative transaction data, accepting increased system complexity as necessary to achieve high-speed automated processing and improved productivity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8041632B1Method and system for using a Bayesian belief network to ensure data integrity

Publication Date: 2011.10.18 CITIBANK N A

AI summary

The present invention relates to a method and system for assessing the risks and/or exposures associated with financial transactions using various statistical and probabilistic techniques. Specifically, the present invention relates to a method and system for identifying plausible sources of error in data used as input to financial risk assessment systems using Bayesian belief networks as a normative diagnostic tool to model relationships between and among inputs/outputs of the risk assessment system and other external factors.