摘要根据相异事物之间的具有相似特点的这一特性来进行区分和归纳分类的过程便称之为聚类。整个聚类执行过程中是无指导人指导的，因此是一种无监视性质的归类。使用数学的方法考虑并解决既定对象的区分归类的方式被称之为聚类分析。原始的聚类分析具有极大限制规范的划分等级，它把要实施认知处理的对象实施十分严格的划分并归属到某一类中，被归类的对象都具有一定相对的的特点，即该分类的类型具有明确的限制。但现实情况往往是很多的对象不具有特定的性质，他们的行为在课堂之上，存在中介，应该恰当实施软划分。

数据掘客与知识的发现是分析聚类的重要的核心方式，本文主要运用模糊理论分析了时间序列，并设计了基于模糊聚类的时间序列分割算法，分析数据的特点，来使其现实具备模糊处理能力，这种方法应用范围极广，并用Matlab进行相应的仿真。分析模糊聚类算法的一般囊括对数据进行标准化、创建模糊相似聚类、矩阵这样的几个步骤。基于模糊聚类的时间序列分割算法具有很好的发展前景。77706

该论文有图6幅，表2个，参考文献20篇。

毕业论文关键词：时间序列分割  模糊聚类  模糊集合

Time Series Segmentation Algorithm Based on Fuzzy Clustering

Abstract According to the characteristic of different things,the process of distinguishing and classifying is called clustering 。Throughout the implementation process of the cluster is no guide, so it is a kind of unsupervised nature of the classification。The way of using mathematical methods to consider and solve the classification of the established objects is called cluster analysis。Original clustering analysis has greatly limited specification grading。It is the implementation of the implementation of cognitive processing of the object is very strict pision and belonging to a class, were classified in terms of the object has a certain relative characteristics, that is the type of the classification has clear limitations。But the reality is often a lot of objects do not have the specific nature of their behavior in the classroom, there is an intermediary, it should be appropriate to implement the soft pision。

Data mining and knowledge discovery is the important core of clustering analysis。By using fuzzy theory this paper mainly analysis the time series, and is designed based on fuzzy clustering algorithm for segmenting time series data analysis, to enable the reality with fuzzy processing capabilities, the scope of application of this method very wide, and the corresponding simulation is carried out with MATLAB。 Analysis of fuzzy clustering algorithm is generally included in the standardization of the data, to create a fuzzy similar clustering, a few steps of the matrix。 The time series segmentation algorithm based on fuzzy clustering has a good development prospect。

There are 6 figures, 2 tables, 20 references in this article。

Key words:Time Series Segmentation  Fuzzy Clustering  the Fuzzy Set

目录

摘  要 I

Abstract II

目录 III

图清单 V

表清单 V

1 绪论 1

1。1课题研究意义与发展现状 1

1。2聚类分析的意义 2

1。3本论文的主要工作 3

2 模糊聚类相关概念 4

2。1模糊聚类理论基础