时态信息检索研究综述

发布时间：2018-04-03 09:27

本文选题：时态信息检索　切入点：时态信息　出处：《数据分析与知识发现》2017年01期

【摘要】：【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。
[Abstract]:[objective] to summarize the present situation of temporal information retrieval at home and abroad,In order to provide a theoretical basis for the scholars concerned to better grasp the research of temporal information retrieval. [literature scope] in Google Scholar, literature retrieval is carried out in terms of "Temporal Information" and "temporal information", respectively, and within an unlimited time range.After obtaining part of the relevant literature, combined with the tracing method, 92 related documents were obtained. [methods] based on the methods of literature investigation and induction, temporal information was extracted from the document, respectively.The research on temporal information retrieval is reviewed and reviewed in the aspects of temporal information recognition and temporal perception ranking. [results] the research shows that the research of temporal information retrieval has the following problems and challenges: foreign time retrievalThere are many researches on state retrieval.However, there are few related studies in China, insufficient research on the attention time of entity and event evolution information identification documents, lack of intention prediction for aperiodic change query.The repeatability of temporal information retrieval model experiment needs to be improved.Document indexing and related applications are reviewed. [conclusion] constructing standardized evaluation data sets and non-parametric temporal information retrieval models will be the future research direction of temporal information retrieval.
【作者单位】：西南大学计算机与信息科学学院;
【基金】：国家社会科学基金青年项目“融合用户个性化与实时性意图的查询推荐模型研究”(项目编号:15CTQ019) 西南大学博士启动基金项目“查询意图自动分类与分析研究”(项目编号:SWU114093)的研究成果之一
【分类号】：G252.7

【相似文献】