Abstract: Because of the unstructured characteristics of free text, text mining becomes an important branch of data mining. In recent years, types of text mining algorithms emerged in large numbers. In this paper, an exploratory text mining algorithm was proposed based on high-dimensional clustering. The algorithm required only a small number of iterations to produce favorable clusters from very large text. Mapping to other recorded data and recording the text to the user group enabled the result of the algorithm be improved further. The feasibility and validity of the proposed method is verified by related data test and the analysis of experimental results.

Key words: free text, high-dimensional clustering, data coverage, text mining, data mining

摘要： 建立了一种基于高维聚类的探索性文本挖掘算法，利用文本挖掘的引导作用实现数据类文本中的数据挖掘。算法只需要少量迭代，就能够从非常大的文本集中产生良好的集群；映射到其他数据与将文本记录到用户组，能进一步提高算法的结果。通过对相关数据的测试以及实验结果的分析，证实了该方法的可行性与有效性。

关键词: 自由文本, 高维聚类, 数据覆盖, 文本挖掘, 数据挖掘