Enhancing cluster labeling using Wikipedia
|Enhancing cluster labeling using Wikipedia|
|Authors:||David Carmel, Haggai Roitman, Naama Zwerdling|
|Citation:||SIGIR '09 Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval : 139-146. 2009 July 19-23. Boston, MA, United states. Association for Computing Machinery.|
|Publication type:||Conference paper|
|Google Scholar cites:||Citations|
|Added by Wikilit team:||Added on initial load|
|Article:||Google Scholar BASE PubMed|
|Other scholarly wikis:||AcaWiki Brede Wiki WikiPapers|
|Web search:||Bing Google Yahoo! — Google PDF|
This work investigates cluster labeling enhancement by utilizing Wikipedia, the free on-line encyclopedia. We describe a general framework for cluster labeling that extracts candidate labels from Wikipedia in addition to important terms that are extracted directly from the text. The labeling quality" of each candidate is then evaluated by several independent judges and the top evaluated candidates are recommended for labeling. Our experimental results reveal that the Wikipedia labels agree with manual labels associated by humans to a cluster much more than with significant terms that are extracted directly from the text. We show that in most cases even when human's associated label appears in the text pure statistical methods have difficulty in identifying them as good descriptors. Furthermore our experiments show that for more than 85% of the clusters in our test collection the manual label (or an inflection or a synonym of it) appears in the top five labels recommended by our system.
"This work investigates cluster labeling enhancement by uti- lizing Wikipedia, the free on-line encyclopedia. We describe a general framework for cluster labeling that extracts candi- date labels from Wikipedia in addition to important terms that are extracted directly from the text. The“labeling qual- ity” of each candidate is then evaluated by several indepen- dent judges and the top evaluated candidates are recom- mended for labeling."
|Topics:||Ranking and clustering systems|
|Theory type:||Design and action|
|Wikipedia coverage:||Sample data|
|Data source:||Archival records, Experiment responses, Wikipedia pages|
|Collected data time dimension:||Cross-sectional|
|Unit of analysis:||Article|
|Wikipedia data extraction:||Dump|
|Wikipedia page type:||Article|
|Wikipedia language:||Not specified|
"Cluster labeling withWikipedia is extremely successful, as shown by our results, especially in collections of documents whose topics are covered well by Wikipedia concepts. For domain specific collections, with topics that are not com- pletely covered by Wikipedia, the proposed candidates may hurt the system’s performance due to their irrelevance to the documents’ topics. For such collections, an intelligent decision should be made regarding the use of Wikipedia or another external resource; alternatively, a choice could be made to focus only on inner terms for labeling. The deci- sion should be made by analyzing the given collection with respect to Wikipedia. Developing such a collection specific decision making as part of the labeling framework is left for further research."
""Cluster labeling withWikipedia is extremely successful, as shown by our results, especially in collections of documents whose topics are covered well by Wikipedia concepts." p. 146"