A multi-party evidence association model construction method and evidence chain extraction method and device

Abstract

The invention provides a Bayesian network-based multi-party evidence association model construction method and evidence chain extraction method and device. These include the construction of evidence network based on fact judgment chain, the calculation of evidence weight and the reasoning method of evidence chain based on Bayesian network. The sources of evidence are divided into plaintiff's evidence, defendant's evidence, judicial appraisal evidence and third-party evidence. First, a multi-party evidence association network is constructed, in which each evidence entity acts as a node in the network, and the correlation probability between nodes in the network is calculated based on the association relationship between evidence elements. Then, based on the evidence types of the event judgment chain, a Bayesian network-based multi-party evidence association model is constructed. Finally, the Bayesian network is optimized by genetic algorithm, and the chain of credible evidence is obtained. The invention can find the most credible evidence chain from multiple sources, and help judicial personnel to screen credible evidence chains from multi-party corroborated or contradictory evidence.

Description

technical field [0001] The invention belongs to the field of artificial intelligence and judicial big data, and specifically relates to a Bayesian network-based multi-party evidence association model construction method and evidence chain extraction method and device. Background technique [0002] As the public's awareness of the rule of law increases, the judicial needs of the people are diversified, and new requirements are put forward for the comprehensiveness of information, the sense of distance in communication, and the timeliness of interaction. Courts at all levels are facing enormous pressure on litigation services, and limited resources are difficult to meet all the demands of different groups of people. Providing convenient litigation services for the public is an important function of the construction of "smart courts", and the review of evidence from multiple parties in a case and the determination of evidence chains are the core of litigation services. [0003...

AI Technical Summary

Problems solved by technology

Many reasoning problems are based on the method of rule judgment, and the rule set is usually complex and huge, requiring manual construction and dynamic maintenance

At the same time, rule-based methods cannot support dynamic and ambiguous reasoning scenarios

In the judicial litigation field, the evidence is heterogeneous and diverse, the sources of evidence are diverse, and the relationship between evidence is complex.

Rule-based methods are difficult to effectively deal with the challenge of judicial evidence relationship modeling

Images