Policy Similarity Evaluation for Accurate Conditional Branch Planning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing policy evaluation systems, such as insurance medical data analysis systems, rely on predictive model information rather than actual values, leading to difficulties in improving evaluation accuracy for inexperienced policies.

Innovation Solution

A policy evaluation method that calculates a similarity measure between a first policy and multiple stored policies, using a processor to output an evaluation value for similar policies with a threshold-based similarity measure, thereby enhancing the accuracy of policy evaluation based on actual values.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If policy evaluation uses predictive model information from insurance medical data analysis systems, then it is possible to generate prediction values for policy effects, but the evaluation accuracy cannot be improved because it does not reach evaluation based on actual values

Engineering Contradiction:

Improveevaluation accuracyVSAvoidlack of actual value information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent creates a virtual copy of the policy evaluation process by constructing a policy flow graph that replicates the structure and logic of actual policy implementation. This virtual model allows evaluation based on simulated actual values rather than purely predictive models, thereby improving measurement precision while avoiding the need for real-world experimentation

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent introduces a policy flow graph as an intermediary between predictive model information and actual policy evaluation. This intermediary structure processes and transforms predictive data into evaluable metrics that approximate actual values, bridging the gap between prediction and real-world outcome measurement

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20260044904A1Policy evaluation method, information processing apparatus, and non-transitory computer-readable recording medium

Publication Date: 2026.02.12 FUJITSU LTD

AI summary

A policy evaluation method includes, when acquiring a first policy that is a policy including a conditional branch and a node connected by a directed edge, referring to a storage that stores a plurality of second policies and an evaluation value of each of the plurality of second policies, the evaluation value indicating a number of targets to which a plurality of targets is assigned to a node by the policy, calculating a similarity measure between the first policy and each of the plurality of second policies, and outputting an evaluation value of a similar policy that is a policy including the conditional branch and the node having the similarity measure equal to or greater than a threshold among the plurality of second policies, by a processor.