首页> 外文会议>International Conference on Computer Science and Education >Thematic Learning-based Full-text Retrieval Research on British and American Journalistic Reading
【24h】

Thematic Learning-based Full-text Retrieval Research on British and American Journalistic Reading

机译:基于专题学习的英美新闻阅读全文检索研究

获取原文

摘要

As for Journalistic Reading Course teaching, it is rather difficult to retrieve instructive and valuable ones from massive online news. In combination with the actual course requirements, the paper endeavors to adopt thematic learning as a means and attach more importance to such three weight indicators as news title, length and timeliness to redesign weight function on the basis of Lucene full-text retrieval algorithm. The comparative experiments prove that the respective addition of length weight, title weight and timeliness weight guarantees the retrieval precision ratio of the top ten improved by 43.6%, 69.2% and 35.9% than before, and by 94.9% after a simultaneous addition of these three weights. It verifies that the search result of the top ten after improvement is more in line with actual teaching requirements in terms of news length and timeliness.
机译:至于新闻阅读课程的教学,要从大量的在线新闻中检索具有指导意义和有价值的内容是相当困难的。结合实际课程要求,本文尝试以专题学习为手段,更加重视新闻标题,时长和时效性三个权重指标,在Lucene全文检索算法的基础上重新设计权重函数。对比实验证明,分别增加长度权重,标题权重和及时性权重,可以确保前十名的检索精度比以前提高了43.6%,69.2%和35.9%,而同时增加了这三种,则提高了94.9%。重量。验证了改进后的前十名的搜索结果在新闻时长和时效性方面更符合实际的教学要求。

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