WordNet、CCD等语义辞典作为自然语言处理的基础工具，被广泛地应用于文本分析、机器翻译等领域。然而，随着Web2.0的到来，各种新的词汇，特别是网络新词日新月异，使得传统的、专家编辑的语义辞典已经跟不上时代的步伐。本项目将研究如何对多个在线百科进行整合，从中提取出活跃的新词，添加到传统的语义辞典中，形成一个全新的、词汇覆盖率更大的语义辞典。这样一方面可提高用户使用语义辞典进行文本分析等应用时的使用效果，同时在专家对语义辞典进行丰富完善时，能够给予一定的帮助，提高精度和效率。除了提出一套通用方法和生成相应的增强版语义辞典外，本项目还考虑如何提炼语义辞典中词汇之间的语义关系，更好地应用于文本分析和机器翻译等任务中。

As a basic tool in natural language processing, semantic lexicons such as WordNet and CCD have widely used in text analysis, machine translation, and other fields. However, along with Web 2.0, various of new words come forth every day. Hence the traditional expert-edited lexical databases cannot keep pace with the latest vocabulary in the real world. This project aims to integrate multiple different online encyclopedias and merge them to WordNet or WordNet-like semantic lexicons. This will not only help to improve the effect and efficiency of applications such as text analysis, but also remarkably help experts to polish the semantic lexicons. This project also plans to propose some learning approaches which can significantly improve the performance of text analysis and machine translation by exploiting the semantic relations in the lexicons.