首页> 外文会议>International Conference on Computer Science and Systems Engineering >Question retrieval based on a conceptual object model
【24h】

Question retrieval based on a conceptual object model

机译:基于概念对象模型的问题检索

获取原文
获取外文期刊封面目录资料

摘要

Question retrieval is an important module of a question answering system. This paper proposes a conceptual object model and uses it to represent domain question, then defines a method of question similarity calculation on the basis of the model, which divides the calculation of question similarity into four parts the calculation of object similarity; the calculation of attribute similarity; the calculation of relation similarity and the calculation of the rest, finally gets the calculation of question similarity by weighted sum the four parts and uses it to retrieve question. Experiments show the new approach is effective to the question retrieval and can improve the accuracy and out-performs the conventional question retrieval approaches.
机译:问题检索是问题应答系统的重要组成部分。本文提出了一个概念对象模型,并使用它来表示域问题,然后根据模型定义一个问题相似性计算的方法,该模型将问题相似性的计算分为四个部分的计算物体相似度的四个部分;属性相似性的计算;关系相似性和其余的计算的计算最终通过加权和四部分来计算相似性的问题,并使用它来检索问题。实验表明,新方法是有效的问题检索,可以提高准确性和外出的传统问题检索方法。

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