With the popularity of Internet and large-scale improvement in the level of enterprise information, the explosive growth of resources, the research of text mining, information filtering and information search appear the unprecedented prospect. So, the cluster technology is becoming the core of text information mining technologies. Clustering is an important form of data mining. This paper introduces common text clustering algorithms, analyses and compares some aspects of clustering algorithms which contains the applicable scope, the initial parameters, termination conditions and noise sensitivity. Algorithms contain hierarchical clustering, partitioned clustering, density-based algorithm and self-organizing maps algorithm.
展开▼
