Journal of Computer Applications ›› 2018, Vol. 38 ›› Issue (3): 620-625.DOI: 10.11772/j.issn.1001-9081.2017092251

Previous Articles     Next Articles

Temporal semantic understanding for intelligent service systems

JIA Shengbin, XIANG Yang   

  1. College of Electronic and Information Engineering, Tongji University, Shanghai 201804, China
  • Received:2017-09-19 Revised:2017-10-16 Online:2018-03-10 Published:2018-03-07
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (71571136), the National Basic Research Program (973 Program) of China (2014CB340404), the Basic Research Project of Science and Technology Commission of Shanghai Municipality (16JC1403000).

面向智能服务系统的时间语义理解

贾圣宾, 向阳   

  1. 同济大学 电子与信息工程学院, 上海 201804
  • 通讯作者: 向阳
  • 作者简介:贾圣宾(1994-),男,山东泰安人,博士研究生,CCF会员,主要研究方向:自然语言处理、服务计算;向阳(1962-),男,上海人,教授,博士,CCF会员,主要研究方向:自然语言处理、服务计算。
  • 基金资助:
    国家自然科学基金资助项目(71571136);国家973计划项目(2014CB340404);上海市科委基础研究项目(16JC1403000)。

Abstract: Aiming at the problem that it is hard to process the temporal semantic information during formulating and providing intelligent services, a temporal semantic understanding model for intelligent service systems was proposed. For service message texts in natural language, temporal information extraction, mapping, and semantic modeling were implemented, so as to provide a universal temporal semantic expression pattern for intelligent service systems. Firstly, a heuristic strategy was adopted to automatically extract temporal phrases and construct time information knowledge base without any manual intervention. Then, a temporal information mapping method based on temporal unit was proposed, to complete quantitative expression of absolute time and logical reasoning of relative time. Finally, a temporal semantic model was constructed by comprehensively using temporal information and contextual information. In service message test set, experimental results show that the precision of time information extraction is as high as 97.58% and the mapping precision is greater than 85%. And the satisfying effect of semantic modeling is shown.

Key words: intelligent service system, temporal semantic understanding, automatic extraction, temporal information mapping

摘要: 针对智能服务制定与提供过程中时间语义处理难的问题,提出一种面向智能服务系统的时间语义信息理解模型。在自然语言描述的服务消息文本上,实现对时间信息的抽取、映射和语义建模,从而为一般的智能服务系统提供通用的时间语义表达模式。首先,模型采用启发式策略自动抽取时间短语并构建时间信息知识库,无需人工干预;然后,提出一种基于时间基元的时间信息映射方法,实现了绝对时间的量化表达以及相对时间的逻辑推理;最后,综合利用时间信息与上下文信息构建时间语义模型。实验结果表明,该模型在服务自然语言文本测试集上,时间信息抽取准确率高达97.58%,时间信息映射准确率高于85%,语义建模效果良好。

关键词: 智能服务系统, 时间语义理解, 自动抽取, 时间信息映射

CLC Number: