Hierarchical Workflow Risk Score Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for performing risk score predictions using multi-party communication data are inefficient and lack reliability, particularly in generating hierarchical-workflow risk scores, due to computational and storage resource intensiveness and the need for extensive training data transmission.

Innovation Solution

The implementation of a hybrid space classification machine learning model and a hybrid-class-based risk scoring model to generate a hybrid class for multi-party communication transcript data objects, followed by a hierarchical risk score adjustment workflow with multiple workflow layers for determining a hierarchical-workflow risk score, which reduces computational and storage needs and improves transmission efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional risk score prediction methods are used with multi-party communication data, then comprehensive risk assessment can be performed, but computational resource consumption and storage requirements increase significantly

Engineering Contradiction:

Improverisk score prediction reliabilityVSAvoidcomputational resource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent segments the risk score prediction process into multiple workflow layers (L layers), where each layer processes specific aspects of the communication data. This segmentation allows the system to handle complex multi-party communication data through divided, manageable stages, reducing the computational burden on any single processing unit while maintaining comprehensive risk assessment capability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a hierarchical dimension to the risk score prediction by implementing an L-layer workflow structure. Each layer operates at a different level of the hierarchy, processing data with varying degrees of complexity. This dimensional transformation allows the system to manage computational resources more efficiently by distributing processing across multiple hierarchical levels rather than concentrating all computation in a single layer.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Reliability

If traditional risk score prediction methods are used with multi-party communication data, then comprehensive risk assessment can be performed, but data transmission requirements increase

Engineering Contradiction:

Improverisk score prediction reliabilityVSAvoiddata transmission volume

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts and processes communication data locally at each workflow layer rather than transmitting all raw multi-party communication data centrally. Each layer extracts relevant features and risk indicators from the communication transcripts locally, significantly reducing the volume of data that needs to be transmitted across the network while maintaining the ability to perform comprehensive risk assessment.

Inventive Principle:

Principle #2Taking out (Extraction)

3Reliability

If hierarchical workflow with multiple layers is implemented, then risk score prediction reliability improves, but system complexity increases

Engineering Contradiction:

Improvehierarchical-workflow risk score predictionVSAvoidworkflow system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements universal processing logic across all L workflow layers, where each layer performs similar risk assessment functions using standardized procedures. This multi-functionality approach allows the same processing framework to be applied repeatedly across different layers, reducing the need for custom complex logic at each layer and thereby reducing overall system complexity while maintaining hierarchical benefits.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Measurement precision

If extensive training data is transmitted for model training, then prediction accuracy improves, but transmission time and resource usage increase

Engineering Contradiction:

Improveprediction accuracyVSAvoiddata transmission time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies local quality processing by allowing each workflow layer to process and analyze communication data locally with its own specialized logic optimized for specific risk indicators. This local processing capability means that training data can be processed distributedly at each layer rather than requiring centralized transmission of all training data, reducing transmission time while maintaining prediction accuracy through localized specialized processing.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS12061639B2Machine learning techniques for hierarchical-workflow risk score prediction using multi-party communication data

Publication Date: 2024.08.13 OPTUM SERVICES IRELAND LTD

AI summary

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing predictive data analysis operations. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform predictive data analysis operations by generating a hybrid class for a multi-party communication transcript data object associated with a predictive entity utilizing a hybrid space classification machine learning model, generating a machine learning-based risk score utilizing a hybrid-class-based risk scoring machine learning model, and generating a hierarchical-workflow risk score using a hierarchical risk score adjustment workflow.