Evaluating Co-Authorship Networks in Author Name Disambiguation for Common Names

摘要

With the increasing size of digital libraries it has become a challenge toidentify author names correctly. The situation becomes more critical whendifferent persons share the same name (homonym problem) or when the names ofauthors are presented in several different ways (synonym problem). This paperfocuses on homonym names in the computer science bibliography DBLP. The goal ofthis study is to evaluate a method which uses co-authorship networks andanalyze the effect of common names on it. For this purpose we clustered thepublications of authors with the same name and measured the effectiveness ofthe method against a gold standard of manually assigned DBLP records. Theresults show that despite the good performance of implemented method for mostnames, we should optimize for common names. Hence community detection wasemployed to optimize the method. Results prove that the applied method improvesthe performance for these names.