首页 | 本学科首页   官方微博 | 高级检索  
     

面向综合集成研讨环境的主动信息获取方法*
引用本文:倪娜,刘凯,李耀东.面向综合集成研讨环境的主动信息获取方法*[J].计算机应用研究,2010,27(11):4058-4062.
作者姓名:倪娜  刘凯  李耀东
作者单位:中国科学院自动化研究所复杂系统与智能科学重点实验室,北京,100190
基金项目:国家“973”重点基础研究发展计划资助项目(2007CB311007)
摘    要:针对在综合集成研讨环境中,由于存在时间压力,传统的网络信息获取方法难以直接使用,提出了一种面向综合集成研讨环境的主动信息获取方法。该方法将领域词条与通用词条相结合,从发言文本流中实时提取话题,并在话题发生变化时自动生成检索词送入搜索引擎进行检索,再通过多个用户之间的协作推荐实现对重要检索结果的筛选。实验结果表明,这种方法可为综合集成研讨系统的用户提供及时、准确、上下文相关的信息服务。

关 键 词:综合集成研讨厅  领域词典  话题提取  协作过滤

Active information retrieval method for CWME
NI N,LIU Kai,LI Yao-dong.Active information retrieval method for CWME[J].Application Research of Computers,2010,27(11):4058-4062.
Authors:NI N  LIU Kai  LI Yao-dong
Affiliation:(Key Laboratory of Complex Systems & Intelligence Science, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China)
Abstract:In CWME, the severe time pressure makes it impossible for participants to retrieve topic-related information timely and sufficiently from the Internet on traditional approaches. To meet this challenge, this paper proposed an active information retrieval method, which could extract topics online and judge change of topics during the discussion process.Once sensed a new topic,query terms generated from this new topic would be sent to a WWW search engine automatically.Identified the valuable search results from the WWW search engine based on collaborative filtering and then sent to the participants actively. Experimental results show that the proposed method is effective in CWME,and can provide timely, accurate and topic-related information for participants well.
Keywords:class imbalance  customer churn prediction  machine learning  sampling
本文献已被 万方数据 等数据库收录!
点击此处可从《计算机应用研究》浏览原始摘要信息
点击此处可从《计算机应用研究》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号