WINGNUS: Keyphrase Extraction Utilizing Document Logical Structure

Proceedings of the 5th International Workshop on Semantic Evaluation, ACL 2010, pages 166–169, Uppsala, Sweden, 15-16 July 2010. c⃝2010 Association for Computational Linguistics WINGNUS: Keyphrase Extraction Utilizing Document Logical Structure Thuy Dung Nguyen Department of Computer Science School of Computing National University of Singapore nguyen14@comp.nus.edu.sg Minh-Thang Luong Department of Computer Science School of Computing National University of Singapore luongmin@comp.nus.edu.sg Abstract We present a system description of the WINGNUS team work1 for the SemEval- 2010 task #5 Automatic Keyphrase Ex- traction from Scientific Articles. A key feature of our system is that it utilizes an inferred document logical structure in our candidate identification process, to limit the number of phrases in the candidate list, while maintaining its coverage of impor- tant phrases. Our top performing system achieves an F1 of 25.22% for the com- bined keyphrases (author and reader as- signed) in the final test data. We note that the method we report here is novel and or- thogonal from other systems, so it can be combined with other techniques to poten- tially achieve higher performance. 1 Introduction Keyphrases are noun phrases (NPs) that capture the primary topics of a document. While benefi- cial for applications such as summarization, clus- tering and indexing, only a minority of documents have manually-assigned keyphrases, as it is a time- consuming process. Automatic keyphrase genera- tion is thus a focus for many researchers. Most existing keyphrase extraction systems view this task as a supervised classification task in two stages: generating a list of candidates – can- didate identification; and using answer keyphrases to distinguish true keyphrases – candidate selec- tion. The selection model uses a set of features that capture the saliency of a phrase as a keyphrase. A major challenge of the keyphrase extraction task lies in the candidate identification process. A narrow candidate list will overlook some true 1This work was supported by a National Research Foun- dation grant “Interactive Media Search” (grant # R-252-000- 325-279). keyphrases (favoring precision), whereas a broad list will produce more errors and require more pro- cessing in latter selection stage (favoring recall). In our previous system (Nguyen and Kan, 2007), we made use of the document logical struc- ture in the proposed features. The premise of this earlier work was that keyphrases are distributed non-uniformly in different logical sections of a pa- per, favoring sections such as introduction, and related work. We introduced features indicating which sections a candidate occurrs in. For our fielded system in this task (Kim et al., 2010), we further leverage the document logical structure for both candidate identification and selection stages. Our contributions are as follows: 1) We suggest the use of Google Scholar-based crawler to auto- matically find PDF files to enhance logical struc- ture extraction; 2) We provide a keyphrase distri- bution study with respect to different logical struc- tures; and 3) From the study result, we propose a candidate identification approach that uses logical structures to effectively limit the number of candi- dates considered while ensuring good coverage. 2 …

Content truncated to 500 words