A Method of Accounting Bigrams in Topic Models
Michael Nokel, Natalia Loukachevitch · 2015
The paper describes the results of an empir-ical study of integrating bigram collocations and similarities between them and unigrams into topic models. First of all, we propose a novel algorithm PLSA-SIM that is a mod-ification of the original algorithm PLSA. It incorporates bigrams and maintains relation-ships between unigrams and bigrams based on their component structure. Then we analyze a variety of word association measures in or-der to integrate top-ranked bigrams into topic models. All experiments were conducted on four text collections of different domains and languages. The experiments distinguish a sub-group of tested measures that produce top-ranked bigrams, which demonstrate signifi-cant improvement of topic models quality for all collections, when integrated into PLSA-SIM algorithm. 1