期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2017
卷号：6
期号：12
页码：22489
DOI：10.15680/IJIRSET.2017.0612049
出版社：S&S Publications
摘要：The ability to determine the topic of a large set of text documents using relevant keywords is usuallyregarded as a very tedious task if done by hand. The problem of keyword extraction from conversations, with the goalof using these keywords to fetch, for each short conversation fragment, a small number of potentially relevantdocuments, which can be recommended to participants, is addressed. A proposed algorithm is to first extract keywordsfrom the conversation which makes use of topic modeling techniques. The keywords which have highest similarity aretaken as keywords. Then, a technique to derive multiple topically separated queries from this keyword set, in order tomaximize the chances of making at least one related recommendation when using these queries to search over theEnglish Wikipedia. The keywords extracted by the algorithm are highly accurate and fit the cluster topic. Keywords areextracted from documents to classify the documents.
关键词：keyword extraction; Document recommendation; topic modeling; clustering; text classification.
