计算机应用研究2025，Vol.42Issue(9)：2614-2620,7.DOI:10.19734/j.issn.1001-3695.2025.02.0036
考虑双重更新的简约扩展置信规则推理决策方法
Simple extended belief rule inference and decision-making method considering dual updating
摘要
Abstract
The continuous accumulation of massive data result in the fact that traditional belief rule inference and decision-making methods are unable to adapt to dynamic decision-making scenarios.This study proposed a simple extended belief rule-based inference and decision-making methodology with the consideration of dual updating.The proposed method used an evi-dential reasoning fusion algorithm to address the redundancy problem in the original extended belief rule-based system by ag-gregating the belief distributions of the consequent attributes under the same antecedent attributes,thereby constructing a sim-ple extended belief rule-based system.The proposed method constructed a two-stage structure updating method by using the greedy algorithm for historical and observed data,and it was implemented for both the original and dynamic rule-based systems.In addition,a Bayesian optimization algorithm updated the parameters of the dynamic rule-based system.As a result,the new approach developed and applied an inference model of the updated extended belief rule for rule inference and decision-making in dynamic scenarios.A practical case of ultrasound-assisted diagnosis of thyroid nodules in the ultrasound department of a ter-tiary hospital in Hefei city,Anhui province was studied to verify the validity and practicality of the proposed method.关键词
简约扩展置信规则库/两阶段结构更新/参数更新/证据推理Key words
simple extended belief rule base/two-stage structure updating/parameter updating/evidential reasoning分类
信息技术与安全科学引用本文复制引用
薛旻,黄浩澜,裴建,侯冰冰..考虑双重更新的简约扩展置信规则推理决策方法[J].计算机应用研究,2025,42(9):2614-2620,7.基金项目
国家自然科学基金资助项目(72001063) （72001063）
中央高校基本科研业务费专项资金资助项目(JZ2023HGTB0282,PA2024GDGP0031) （JZ2023HGTB0282,PA2024GDGP0031）
