Using Genetic Programming to Evaluate the Impact of Social Network Analysis in Author Name Disambiguation

In digital libraries, which have become extremely popular in the scientific community, often people want to find publications by an author using the author name as a query. However, since authors may have many denominations and one denomination may refer to many authors, name searches may present ambiguous results. To tackle this problem, several studies have been developed. Recently the use of social networks has been studied in author name disambiguation. In this article, we use a machine learning approach based on Genetic Programming to evaluate the impact of social network analysis in author name disambiguation. Through experiments using real-world data, we show that social network analysis greatly improves the quality of results. Also, we demonstrate that our approach is able to compete with state-of-the-art techniques.