从数据挖掘偏差到文献计量偏差

发布时间：2018-03-17 12:02

本文选题：数据挖掘　切入点：数据挖掘偏差　出处：《情报理论与实践》2017年10期 　论文类型：期刊论文

【摘要】：[目的/意义]互联网的迅猛发展使得信息量增速不断加快。作为从海量数据中发现有用知识的有力手段,数据挖掘成为近年的研究热点。然而在数据挖掘过程中,输出值和真实值之间往往存在着一定的差距,即数据挖掘偏差。[方法/过程]通过相关研究综述了数据挖掘偏差的概念、研究进展以及发展方向,并以数据挖掘的基本步骤类比文献计量的基本步骤,进而提出文献计量偏差的基本概念。从文献来源选取、文献数据预处理、文献计量方法选取和计量结果解读4个角度重点论述了文献计量偏差的主要表现方式和解决办法。[结果/结论]文章旨在呼吁未来该领域的相关研究关注文献计量偏差及其带来的负向影响,并期望相关研究能够通过科学方法避免文献计量偏差,从而得到更为准确和可靠的结论。
[Abstract]:[objective / significance] the rapid development of the Internet has accelerated the increase in the amount of information. As a powerful means of finding useful knowledge from massive data, data mining has become a research hotspot in recent years. However, in the process of data mining, There is often a gap between the output value and the real value, that is, the deviation of data mining. [method / process] the concept, research progress and development direction of data mining deviation are summarized through relevant studies. The basic steps of data mining are analogous to the basic steps of document metrology, and then the basic concept of document metrology deviation is put forward. This paper mainly discusses the main manifestation and solution of bibliometric deviation from four angles: the selection of bibliometric methods and the interpretation of measurement results. [results / conclusions] this paper aims to appeal to the related research in this field to pay attention to bibliometrics in the future. Volume deviation and its negative effects, It is expected that the relevant research can avoid the metrological deviation through scientific methods, and thus obtain more accurate and reliable conclusions.
【作者单位】：北京大学信息管理系;印第安纳大学信息学与计算机学院;
【分类号】：G353.1

【相似文献】