arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 0912.2307cs.IRcs.DB
收藏

基于秩的聚类在生物医学数据库中的应用

Rank Based Clustering For Document Retrieval From Biomedical Databases

Jayanthi Manicassamy, P. Dhavachelvan

摘要:

如今,搜索引擎被广泛用于从世界各地的各种资源中提取信息。其中,大多数检索位于生物医学领域,用于从各种生物医学数据库中检索相关文献。目前的搜索引擎缺乏对文档的聚类和对文档相关性的表征。为了克服这些缺陷,基于文本的搜索引擎已被开发用于从Medline和PubMed生物医学数据库中检索文档。该搜索引擎引入了基于聚类的网页排序概念,自动表示聚类基础上的相关性。除此之外,图树构造用于表示联网在一起的文档的相关性水平。基于搜索引擎的生物医学文档的这种高级功能整合发现在基于相关性的审查相关文档中提供更好的结果。

英文摘要:

Now a day's, search engines are been most widely used for extracting information's from various resources throughout the world. Where, majority of searches lies in the field of biomedical for retrieving related documents from various biomedical databases. Currently search engines lacks in document clustering and representing relativeness level of documents extracted from the databases. In order to overcome these pitfalls a text based search engine have been developed for retrieving documents from Medline and PubMed biomedical databases. The search engine has incorporated page ranking bases clustering concept which automatically represents relativeness on clustering bases. Apart from this graph tree construction is made for representing the level of relatedness of the documents that are networked together. This advance functionality incorporation for biomedical document based search engine found to provide better results in reviewing related documents based on relativeness.

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。

↑