统计学习聚类方法的应用研究摘要聚类分析指将物理或抽象对象的集合分组为由类似的对象组成的多个类的分析过程 .聚类分析作为一种有效的数据分析手段，能够帮助人们认识和揭示事物之间的内在联系，它已被广泛应用到数据压缩、图像处理、计算机视觉、文本聚类和异常点检测等领域.本文首先展示了统计学习的理论基础，然后详细介绍了 k-均值法、基于图的谱聚类、自组织神经网络、层次聚类等聚类方法，最后使用 k-均值算法来判断中国乒乓球水平以及利用谱聚类来实现图像分割等应用．关键词：k-均值法；谱聚类；自组织学习；层次聚类 Research on Application of Statistical Learning Clustering MethodABSTRACTCluster analysis refers to the analysis process of grouping a set of physical or abstract objects into multiple classes composed of similar objects. As an effective means of data analysis, clustering analysis can reveal the internal relationship between things. It has been widely used in data compression, image processing, computer vision, text clustering, outlier detection and other fields. This paper first shows the theoretical basis of statistical learning, then introduces the k-means method, spectrum clustering based on graph, self-organizing neural network, hierarchical clustering and other clustering methods in detail, and finally uses k-means algorithm to judge the level of Chinese table tennis and uses spectrum clustering to achieve image segmentation.Keywords： K-means; Spectral Clustering; Self-Organized Learning; Hierarchical ClusteringI目录摘要.......................................................................................................................................................IABSTRACT.........................................................................................................................................II1 研究意义和目的..............................................................................................................................
