中文文本挖掘基本理论与应用.pdfVIP

中文文本挖掘基础理论与应用 摘要 中文文本挖掘基础理论与应用 摘要 文本挖掘的应用前景十分广泛，中文文本较西方语言文本有其独特性，本文 研究的对象是中文文本。此文首先对文本挖掘的背景和发展简要综述，并介绍中 文文本挖掘的概念和流程，其中重点介绍了特征的提取、降维及分类算法。介绍 了R 中文本挖掘相关的包，以及CHQS 文本多分类系统，并应用此系统对文档 进行分类。 本文重点是借助被广泛应用的开源工具构建自己的中文文本挖掘系统。根据 中文文本挖掘的流程，详细的描述了自己对系统的探索，利用LTP 系统进行分 词，然后借助Weka 的StringTOWordVector 完成了从非结构化数据到结构化数据 的转化，最后用LibSVM 做中文文本多分类模型的训练和预测。 关键词：文本挖掘、文本分类、中文文本 作者：康东 导师：唐煜 I Abstract Basic Theory of Chinese Text Mining and its Applications Abstract Text mining has an extensive application prospect. Compared with Western Linguistic text, Chinese text has its own uniqueness. Given that reason, this paper selects Chinese text as its study object. First of all, the background and development of text mining are reviewed, and then presents its conception, procedures, and especially the feature extraction, dimensionality reduction and Classification Algorithm. Packages of Chinese text mining in R and CHQS text multi- classified system are introduced, and apply this system to classify documents. This paper focuses on using widely used open source tools to build your own Chinese text mining system. According to the flow of Chinese text mining, it describes the exploration of the system. To segment words with the LTP system, transform unstructured textual data to the structured data using StringTOWordVector of Weka, finally with LibSVM train Chinese text classification model and use it to predict. Keywords: text mining, text classificati