DFKI KeyWE: Ranking Keyphrases Extracted from Scientific Articles

Proceedings of the 5th International Workshop on Semantic Evaluation, ACL 2010, pages 150–153, Uppsala, Sweden, 15-16 July 2010. c⃝2010 Association for Computational Linguistics DFKI KeyWE: Ranking keyphrases extracted from scientific articles Kathrin Eichler DFKI - Language Technology Berlin, Germany kathrin.eichler@dfki.de G¨unter Neumann DFKI - Language Technology Saarbr¨ucken, Germany neumann@dfki.de Abstract A central issue for making the content of a scientific document quickly acces- sible to a potential reader is the extrac- tion of keyphrases, which capture the main topic of the document. Keyphrases can be extracted automatically by generating a list of keyphrase candidates, ranking these candidates, and selecting the top-ranked candidates as keyphrases. We present the KeyWE system, which uses an adapted nominal group chunker for candidate ex- traction and a supervised ranking algo- rithm based on support vector machines for ranking the extracted candidates. The system was evaluated on data provided for the SemEval 2010 Shared Task on Keyphrase Extraction. 1 Introduction Keyphrases capture the main topic of the docu- ment in which they appear and can be useful for making the content of a document quickly ac- cessible to a potential reader. They can be pre- sented to the reader directly, in order to provide a short overview of the document, but can also be processed further, e.g. for text summarization, document clustering, question-answering or rela- tion extraction. The task of extracting keyphrases automatically can be performed by generating a list of keyphrase candidates, ranking these can- didates, and selecting the top-ranked candidates as keyphrases. In the KeyWE system, candidates are generated based on an adapted nominal group chunker described in section 3 and ranked using the SVMrank algorithm (Joachims, 2006), as de- scribed in section 4. The used features are spec- ified in section 5. In section 6, we present the results achieved on the test data provided for the SemEval 2010 Shared Task on Keyphrase Extrac- tion1 by selecting as keyphrases the top 5, 10, and 15 top-ranked candidates, respectively. 2 Related work The task of keyphrase extraction came up in the 1990s and was first treated as a supervised learn- ing problem in the GenEx system (Turney, 1999). Since then, the task has evolved and various new approaches have been proposed. The task is usu- ally performed in two steps: 1. candidate ex- traction (or generation) and 2. keyphrase selec- tion. The most common approach towards can- didate extraction is to generate all n-grams up to a particular length and filter them using stopword lists. Lately, more sophisticated candidate extrac- tion methods, usually based on additional linguis- tic information (e.g. POS tags), have been pro- posed and shown to produce better results (e.g. Hulth (2004)). Liu et al. (2009) restrict their can- didate list to verb, noun and adjective words. Kim and Kan (2009) generate regular expression rules to extract simplex nouns and nominal phrases. As the majority of technical terms is in nominal group positions2, we assume that the same holds true for keyphrases and apply an adapted nominal group chunker to extract …

Content truncated to 500 words