UvT: The UvT Term Extraction System in the Keyphrase Extraction Task

Proceedings of the 5th International Workshop on Semantic Evaluation, ACL 2010, pages 194–197, Uppsala, Sweden, 15-16 July 2010. c⃝2010 Association for Computational Linguistics UvT: The UvT Term Extraction System in the Keyphrase Extraction task Kalliopi Zervanou ILK / TiCC - Tilburg centre for Cognition and Communication University of Tilburg, P.O. Box 90153, 5000 LE Tilburg, The Netherlands K.Zervanou@uvt.nl Abstract The UvT system is based on a hybrid, lin- guistic and statistical approach, originally proposed for the recognition of multi- word terminological phrases, the C-value method (Frantzi et al., 2000). In the UvT implementation, we use an extended noun phrase rule set and take into consideration orthographic and morphological variation, term abbreviations and acronyms, and ba- sic document structure information. 1 Introduction The increasing amount of documents in elec- tronic form makes imperative the need for docu- ment content classification and semantic labelling. Keyphrase extraction contributes to this goal by the identification of important and discriminative concepts expressed as keyphrases. Keyphrases as reduced document content representations may find applications in document retrieval, classifica- tion and summarisation (D’Avanzo and Magnini, 2005). The literature distinguishes between two principal processes: keyphrase extraction and keyphrase assignment. In the case of keyphrase assignment, suitable keyphrases from an exist- ing knowledge resource, such as a controlled vo- cabulary, or a thesaurus are assigned to docu- ments based on classification of their content. In keyphrase extraction, the phrases are mined from the document itself. Supervised approaches to the problem of keyphrase extraction include the Naive Bayes-based KEA algorithms (Gordon et al., 1999) (Medelyan and Witten, 2006), deci- sion tree-based and the genetic algorithm-based GenEx (Turney, 1999), and the probabilistic KL divergence-based language model (Tomokiyo and Hurst, 2003). Research in keyphrase extrac- tion proposes the detection of keyphrases based on various statistics-based, or pattern-based fea- tures. Statistical measures investigated focus pri- marily on keyphrase frequency measures, whereas pattern-features include noun phrase pattern filter- ing, identification of keyphrase head and respec- tive frequencies (Barker and Cornacchia, 2000), document section position of the keyphrase (e.g., (Medelyan and Witten, 2006)) and keyphrase coherence (Turney, 2003). In this paper, we present an unsupervised approach which combines pattern-based morphosyntactic rules with a statis- tical measure, the C-value measure (Frantzi et al., 2000) which originates from research in the field of automatic term recognition and was initially de- signed for specialised domain terminology acqui- sition. 2 System description The input documents in the Keyphrase Extrac- tion task were scientific articles converted from their originally published form to plain text. Due to this process, some compound hyphen- ated words are erroneously converted into a single word (e.g., “resourcemanagement” vs. “resource- management”). Moreover, document sections such as tables, figures, footnotes, headers and foot- ers, often intercept sentence and paragraph text. Finally, due to the particularity of the scientific ar- ticles domain, input documents often contain ir- regular text, such as URLs, inline bibliographic references, mathematical formulas and symbols. In our approach, we attempted to address some of these issues by document structuring, treatment of orthographic variation and …

Content truncated to 500 words