%0 Conference Paper %1 DBLP:conf/dexaw/WartenaB08 %A Wartena, Christian %A Brussee, Rogier %B DEXA Workshops %D 2008 %I IEEE Computer Society %K clustering con_ocurance tag topic %P 54-58 %T Topic Detection by Clustering Keywords %X We consider topic detection without any prior knowledge of category structure or possible categories. Keywords are extracted and clustered based on different similarity measures using the induced k-bisecting clustering algorithm. Evaluation on Wikipedia articles shows that clusters of keywords correlate strongly with the Wikipedia categories of the articles. In addition, we find that a distance measure based on the Jensen-Shannon divergence of probability distributions outperforms the cosine similarity. In particular, a newly proposed term distribution taking co-occurrence of terms into account gives best results. %@ 978-0-7695-3299-8

@inproceedings{DBLP:conf/dexaw/WartenaB08, abstract = {We consider topic detection without any prior knowledge of category structure or possible categories. Keywords are extracted and clustered based on different similarity measures using the induced k-bisecting clustering algorithm. Evaluation on Wikipedia articles shows that clusters of keywords correlate strongly with the Wikipedia categories of the articles. In addition, we find that a distance measure based on the Jensen-Shannon divergence of probability distributions outperforms the cosine similarity. In particular, a newly proposed term distribution taking co-occurrence of terms into account gives best results.}, added-at = {2014-03-04T07:10:18.000+0100}, author = {Wartena, Christian and Brussee, Rogier}, bibsource = {DBLP, http://dblp.uni-trier.de}, biburl = {https://www.bibsonomy.org/bibtex/2848c358fab6dae7f498b0a391cc72211/inmantang}, booktitle = {DEXA Workshops}, crossref = {DBLP:conf/dexaw/2008}, ee = {http://doi.ieeecomputersociety.org/10.1109/DEXA.2008.120}, interhash = {c506dc044945fef757211c9eafdd932e}, intrahash = {848c358fab6dae7f498b0a391cc72211}, isbn = {978-0-7695-3299-8}, keywords = {clustering con_ocurance tag topic}, pages = {54-58}, publisher = {IEEE Computer Society}, timestamp = {2014-03-04T07:11:38.000+0100}, title = {Topic Detection by Clustering Keywords}, year = 2008 }