TY - CONF TI - Labelling Topics using Unsupervised Graph-based Methods AU - Aletras, Nikolaos AU - Stevenson, Mark T2 - Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) AB - This paper introduces an unsupervised graph-based method that selects textual labels for automatically generated topics. Our approach uses the topic keywords to query a search engine and generate a graph from the words contained in the results. PageRank is then used to weigh the words in the graph and score the candidate labels. The state-of-the-art method for this task is supervised (Lau et al., 2011). Evaluation on a standard data set shows that the performance of our approach is consistently C1 - Baltimore, Maryland C3 - Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) DA - 2014/// PY - 2014 DO - 10.3115/v1/P14-2103 DP - Crossref SP - 631 EP - 636 LA - en PB - Association for Computational Linguistics UR - http://aclweb.org/anthology/P14-2103 Y2 - 2019/01/10/07:55:29 KW - Topic model labeling KW - Topic modeling ER -