• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International Journal of Innovative Computing Information and Control >AUTOMATIZATION NEWS GROUPING USING LATENT DIRICHLET ALLOCATION FOR IMPROVING EFFICIENCY
【24h】

AUTOMATIZATION NEWS GROUPING USING LATENT DIRICHLET ALLOCATION FOR IMPROVING EFFICIENCY

机译：自动化新闻分组使用潜在的Dirichlet分配提高效率

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

News is one of the most frequently occurring text data in human life. The news appears in the form of printed news or online news. With new news every day. there is a lot of news in various life fields. With that different kind of news, the redactor needs to read them and then divide them into various fields. To group news into their respective fields, people need a lot of time to read the news and group them into the appropriate fields. With this condition, soon the tasks will be piled up. News grouping will also help the reader to find news based on the topic that they want to read. The existing problems inspire the researcher to conduct research about grouping news automatically. This research will use the Latent Dirichlet Allocation (LDA) method with perplexity calculation for choosing the number of the topics that gives a better prediction. The optimal number of topics that found after a comparison of alpha value and perplexity value is twenty-nine topics with a perplexity value of 997.5468 and an alpha value of 0.2. The result of this research is groups of the word that can be concluded into many topics.
机译：新闻是人类生活中最常发生的文本数据之一。该消息以印刷新闻或在线新闻的形式出现。每天新闻。各种生活领域有很多新闻。通过这种不同的新闻，重氮机需要阅读它们，然后将它们分成各种领域。将新闻集团新闻进入各自的领域，人们需要花费大量时间阅读新闻并将其分组到相应的领域。有了这种情况，很快就会堆积。新闻分组还将帮助读者根据他们想要阅读的主题找到新闻。现有问题激发了研究人员，可以自动进行研究新闻。该研究将使用潜在的Dirichlet分配（LDA）方法具有困惑计算来选择提供更好预测的主题的数量。在α值和困惑值比较之后发现的最佳局部数是具有997.5468的困惑值的二十九个主题和0.2的α值。该研究的结果是可以在许多主题中得出结论的单词组。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号