时态信息检索研究综述

来源 :数据分析与知识发现 | 被引量 : 0次 | 上传用户:hxz22
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
【目的】总结国内外时态信息检索研究现状,以期为相关学者更好地把握时态信息检索研究问题提供理论基础。【文献范围】在Google Scholar中分别以检索式“Temporal Information”与“时态信息”且不限定时间范围地进行文献检索,获得部分相关文献后,再结合追溯法最终得到92篇相关文献。【方法】基于文献调研与归纳总结方法,分别从文档中时态信息抽取、查询中时态信息识别和时间感知排序三方面对时态信息检索的相关研究进行综述与评述。【结果】研究发现时态信息检索研究存在着如下问题和挑战:国外对时态检索研究比较多,而国内的相关研究甚少;利用表征时间信息的实体与事件演化信息识别文档关注时间的相关研究不足;缺乏对非周期变化查询的意图预测;时态信息检索模型实验的可重复性有待提高。【局限】未对该领域的文档采集、文档索引以及相关应用进行文献综述。【结论】构建标准化的评测数据集以及无参数时态信息检索模型将是时态信息检索领域的未来方向研究。 【Objective】 The purpose of this paper is to summarize the research status of temporal information retrieval at home and abroad in order to provide a theoretical basis for the scholars to grasp the research questions of temporal information retrieval better. 【Scope】 In the Google Scholar search, “Temporal Information” and “Tense Information” and do not limit the time range for document retrieval, access to some relevant literature, combined with retrospective 92 final articles Related documents 【Method】 Based on the methods of literature review and induction, the related research on temporal information retrieval was summarized from the aspects of temporal information extraction, temporal information recognition and temporal information retrieval. 【Result】 The study found that the research on temporal information retrieval has the following problems and challenges: There are many researches on temporal retrieval in foreign countries, but little domestic research; the use of the entity that characterizes temporal information and the event evolution information to identify the document concerns the time Lack of research; lack of intent forecast for aperiodic change inquiring; the repeatability of temporal information retrieval model experiment needs to be improved. [Limitations] There is no literature review of document collection, document indexing, and related applications in this area. [Conclusion] Constructing a standardized evaluation dataset and a non-parametric temporal information retrieval model will be the future direction of temporal information retrieval.
其他文献