arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1112.2071cs.IR
收藏

语篇语料库的主位分析与可视化

Thematic Analysis and Visualization of Textual Corpus

Anja Habacha Chabi, Ferihane Kboubi, Mohamed Ben Ahmed

摘要:

文档的语义分析是当前研究的热点。这个领域的工作可以采取几个方向,并触及几个粒度级别。在目前的工作中,我们感兴趣的正是文本文件的主题分析。在我们的方法中,我们建议研究主位关联在语篇中的变化,以确定主位和语篇中所唤起的所有次位。这使我们能够在第二个层次的分析,以确定在语篇语料库中的主位关联的关系。通过对这些关联关系的识别和分析,我们建议生成主题路径,允许用户在信息搜索系统的框架内,根据他们感兴趣的主题来探索语料库,并通过导航主题关联关系来发现新知识。

英文摘要:

The semantic analysis of documents is a domain of intense research at present. The works in this domain can take several directions and touch several levels of granularity. In the present work we are exactly interested in the thematic analysis of the textual documents. In our approach, we suggest studying the variation of the theme relevance within a text to identify the major theme and all the minor themes evoked in the text. This allows us at the second level of analysis to identify the relations of thematic associations in a textual corpus. Through the identification and the analysis of these association relations we suggest generating thematic paths allowing users, within the frame work of information search system, to explore the corpus according to their themes of interest and to discover new knowledge by navigating in the thematic association relations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑