基于模糊子空间聚类的0阶岭回归TSK模糊系统  

Fuzzy subspace clustering based 0-order ridge regression TSK fuzzy system

作　　者：邓赵红[1] 张江滨[1] 蒋亦樟[1] 王士同[1] 

机构地区：[1]江南大学数字媒体学院,江苏无锡214122

出　　处：《控制与决策》2016年第5期882-888,共7页Control and Decision

基　　金：国家自然科学基金项目(61170122);江苏省杰出青年基金项目(BK20140001);新世纪优秀人才支持计划项目(NCET120882)

摘　　要：经典数据驱动型TSK模糊系统在利用高维数据训练模型时,由于规则前件采用的特征过多,导致规则的解释性和简洁性下降.对此,根据模糊子空间聚类算法的子空间特性,为TSK模型添加特征抽取机制,并进一步利用岭回归实现后件的学习,提出一种基于模糊子空间聚类的0阶岭回归TSK模型构建方法.该方法不仅能为规则抽取出重要子空间特征,而且可为不同规则抽取不同的特征.在模拟和真实数据集上的实验结果验证了所提出方法的优势.The classical data-driven Takagi-Sugeno-Kang（TSK） fuzzy system extracts more features for structuring the antecedent of the fuzzy rule when trained by high dimensional data, and the interpretation of system is degenerated and the linguistic interpretation is complex. A fuzzy modeling model for the fuzzy subspace clustering based zero-order ridge regression TSK fuzzy system is proposed, in which the feature extraction mechanism based on the subspace feature of fuzzy subspace clustering is added, and the ridge regression is used to realize the learning of consequent. The proposed method not only can extract important features for structuring fuzzy rules, but also can extract different features for different rules. The experimental results on the synthetic and real-world datasets show the advantage of the proposed method.

关 键 词：解释性 高维数据 岭回归 TSK模糊系统 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]