A Comparison of Mapping Algorithms for Author Co-Citation Data Analysis

摘要

A key process of any citation analysis study is to map therncoded citation data from a high-dimensional dataset to arnlower dimensional one while detecting the groups, clusters,rnpatterns or other features of the citation relationships. Overrnthe years, many methods have been used in various studies,rnincluding multi-dimensional scaling, Pathfinder networks,rnKohonen’s self-organizing mapping, etc. Many of thesernmethods are fundamentally different, but their results arernsimilar and comparable. In this study, we selected andrnapplied four of the mapping methods to the same dataset,rnthe author co-citation matrix of the top 100 highly citedrninformation scientists. The results of the different mappingrnmethods provide interesting comparisons among therndifferent mapping algorithms as well as the different viewsrnof the dataset.