用正交最小二乘聚类和模糊辨识系统建立紧凑型TS-模糊模型非线性预测恶唑啉及其衍生物的抗结核活性
Non linear Prediction of Antitubercular Activity Of Oxazolines and Oxazoles derivatives Making Use of Compact TS-Fuzzy models Through Clustering with orthogonal least sqaure technique and Fuzzy identification system
摘要:
不确定非线性预测系统的预测是一个重要而具有挑战性的问题。模糊逻辑模型往往是一个很好的选择来描述这样的系统,但在许多情况下,这些很快变得复杂。通常在描述符选择和创建适当的局部规则方面投入的努力太少。此外,通常不应用模型简化,而这可以通过移除冗余数据来分析模型。本文提出了一种组合方法,处理这些问题,以创建紧凑的高木Sugeno(TS)模型,可以有效地用于表示复杂的预测系统。提出了一种新的模糊聚类方法用于紧致TS模糊模型的辨识。在此基础上,利用正交最小二乘技术选择TS模型的最佳相关后件变量,并提出了一种基于Fisher类间可分性基的前件变量选择方法.这个完整的方法证明了通过恶唑啉和恶唑衍生物作为抗结核药物的非线性回归基准。并与神经模糊算法ANFIS和先进的模糊聚类技术FMID工具箱进行了比较。
英文摘要:
The prediction of uncertain and predictive nonlinear systems is an important and challenging problem. Fuzzy logic models are often a good choice to describe such systems however in many cases these become complex soon. commonlly, too less effort is put into descriptor selection and in the creation of suitable local rules. Moreover, in common no model reduction is applied, while this may analyze the model by removing redundant data. This paper suggests a combined method that deal with these issues in order to create compact Takagi Sugeno (TS) models that can be effectively used to represent complex predictive systems. A new fuzzy clustering method is come up with for the identification of compact TS-fuzzy models. The best relevant consequent variables of the TS model are choosen by an orthogonal least squares technique based on the obtained clusters.For the selection of the relevant antecedent (scheduling) variables a new method has been developed based on Fisher's interclass separability basis. This complete approach is demonstrated by means of the Oxazolines and Oxazoles derivatives as antituberculosis agent for nonlinear regression benchmark. The results are compared with results obtained by neuro-fuzzy i.e. ANFIS algorithm and advanced fuzzyy clustering techniques i.e FMID toolbox .
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。