A review of recent studies on literature-based discovery
Dai Bing,Hu Zhengyin
(Chengdu Library and Information Center, Chinese Academy of Sciences, Chengdu 610041, China)
(Department of Library, Information and Archives Management, School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190, China)
[目的]对基于文献的知识发现(Literature-based discovery,LBD)近十年的文献进行综述,有助于学界了解该主题的新近研究进展、发展趋势与面临挑战。[文献范围]使用“literature based discovery”、“literature AND knowledge discovery”、“文献知识发现”、“文献 AND 知识挖掘”在Web of Science、CNKI和百度学术中检索,限定文献发表时间为2010年-2020年,共筛选出72篇代表性文献进行述评。[方法]从研究对象、方法技术、结果评估与典型应用四个方面对文献进行归纳梳理,并总结LBD的发展趋势与面临挑战。[结果]研究发现LBD发展呈现出研究对象复杂化、分析方法智能化、发现结果丰富化与应用服务实践化的趋势;LBD在多源异构数据融合、知识发现可解释性、结果有效性评估、多领域专家协同方面面临重大挑战。[局限]本文主要基于文献对LBD新近进展进行综述,对LBD工具系统及产业界应用覆盖不够。[结论]作为情报学、信息学、数据科学的交叉研究领域,LBD对挖掘跨学科领域隐性知识与提供高质量学科化知识服务具有重要意义,但距离真正实现支持潜在的科学新发现还存在诸多挑战。
[Objective] This paper investigates the literatures of literature-based discovery (LBD) in the recent ten years, which will help the researchers better understand the new research progress, development trend and challenges on this topic. [Coverage] Using "literature based discovery", "literature and knowledge discovery" in Chinese and English to search in the databases of web of science, CNKI and Baidu academic, the publication year is limited from 2010 to 2020, and 72 representative literatures are selected for review. [Methods] Firstly, this paper summarized the literatures from four aspects: research objects, methods and techniques, result evaluation and typical applications, and after that the future development trend and challenges of LBD are summarized. [Results] The research objects of LBD tend to be more complicated, the analysis methods and techniques tend to be more intelligent, the discovery results are more enriched, and more LBD applications appear. LBD also faces some challenges in multi-source heterogeneous data fusion, interpretability of knowledge discovery, effectiveness evaluation of results, and collaboration of multi-disciplinary experts. [Limitations] This paper mainly reviews the recent development of LBD based on literatures, and it is not enough to cover the LBD tools or systems and industry applications. [Conclusions] As an interdisciplinary research field of information science, informatics and data science, LBD is of great significance for mining interdisciplinary knowledge and providing high-quality subject knowledge services. However, there are still some challenges to support potential scientific discoveries.
代冰, 胡正银.
基于文献的知识发现新近研究综述
[J]. 数据分析与知识发现, 10.11925/infotech.2096-3467.2020.1155.
Dai Bing, Hu Zhengyin.
A review of recent studies on literature-based discovery
. Data Analysis and Knowledge Discovery, 0, (): 1-.