A principled methodology for comparing relatedness measures for clustering publications

There are many different relatedness measures, based for instance on citation relations or textual similarity, that can be used to cluster scientific publications. We propose a principled methodology for comparing relatedness measures and for evaluating the accuracy of clustering solutions resulting from these measures. Our empirical results, based on publications in the field of condensed matter physics, suggest that bibliographic coupling relations yield more accurate clustering solutions than direct citation relations and co-citation relations. The so-called extended direct citation approach is competitive with bibliographic coupling in terms of the accuracy of the resulting clustering solutions. Our results also suggest that BM25 yields more accurate clustering solutions than other text-based relatedness measures.