计算机应用研究2025，Vol.42Issue(11)：3355-3362,8.DOI:10.19734/j.issn.1001-3695.2025.03.0072
基于规则插值的模糊模式树优化方法
Optimization method of fuzzy pattern trees based on rule interpolation
摘要
Abstract
PTTD uses similarity measures to induce pattern trees.During model construction,it adopts a beam search strategy and explores the candidate tree space greedily,resulting in prolonged model-building time.To address this issue,this paper proposed an optimized PTTD strategy based on fuzzy rule interpolation(PTTD-FRI).Firstly,this paper designed a new search strategy to simplify the search space and construct a sparse rule base.Secondly,this paper introduced a transformation approach to convert the complex fuzzy pattern tree into multiple fuzzy rules containing only t-norm,which were aggregated using t-conorm.Finally,a fuzzy rule interpolation module further refined the rule base and improved classification accuracy through simple computations.The experimental results show that PTTD-FRI effectively improves the training efficiency and re-duces the model complexity while maintaining comparable classification accuracy.关键词
模式树/模糊规则插值/聚合算子/分类/稀疏规则库Key words
pattern tree/fuzzy rule interpolation/aggregation operator/classification/sparse rule base分类
计算机与自动化引用本文复制引用
苏攀,任雪莹,梅华威..基于规则插值的模糊模式树优化方法[J].计算机应用研究,2025,42(11):3355-3362,8.基金项目
中央高校基本科研业务费专项资金资助项目(2024MS129) （2024MS129）
河北省高等学校科学技术研究项目(QN2023181) （QN2023181）
