I2R: Three Systems for Word Sense Discrimination, Chinese Word Sense Disambiguation, and English Word Sense Disambiguation

Proceedings of the 4th International Workshop on Semantic Evaluations (SemEval-2007), pages 177–182, Prague, June 2007. c⃝2007 Association for Computational Linguistics I2R: Three Systems for Word Sense Discrimination, Chinese Word Sense Disambiguation, and English Word Sense Disambiguation Zheng-Yu Niu, Dong-Hong Ji Institute for Infocomm Research 21 Heng Mui Keng Terrace 119613 Singapore niu zy@hotmail.com dhji@i2r.a-star.edu.sg Chew-Lim Tan Department of Computer Science National University of Singapore 3 Science Drive 2 117543 Singapore tancl@comp.nus.edu.sg Abstract This paper describes the implementation of our three systems at SemEval-2007, for task 2 (word sense discrimination), task 5 (Chinese word sense disambiguation), and the first subtask in task 17 (English word sense disambiguation). For task 2, we ap- plied a cluster validation method to esti- mate the number of senses of a target word in untagged data, and then grouped the in- stances of this target word into the esti- mated number of clusters. For both task 5 and task 17, We used the label propagation algorithm as the classifier for sense disam- biguation. Our system at task 2 achieved 63.9% F-score under unsupervised evalua- tion, and 71.9% supervised recall with su- pervised evaluation. For task 5, our sys- tem obtained 71.2% micro-average preci- sion and 74.7% macro-average precision. For the lexical sample subtask for task 17, our system achieved 86.4% coarse- grained precision and recall. 1 Introduction SemEval-2007 launches totally 18 tasks for evalua- tion exercise, covering word sense disambiguation, word sense discrimination, semantic role labeling, and sense disambiguation for information retrieval, and other topics in NLP. We participated three tasks in SemEval-2007, which are task 2 (Evaluating Word Sense Induction and Discrimination Systems), task 5 (Multilingual Chinese-English Lexical Sam- ple Task) and the first subtask at task 17 (English Lexical Sample, English Semantic Role Labeling and English All-Words Tasks). The goal for SemEval-2007 task 2 (Evaluat- ing Word Sense Induction and Discrimination Sys- tems)(Agirre and Soroa, 2007) is to automatically discriminate the senses of English target words by the use of only untagged data. Here we address this word sense discrimination problem by (1) estimat- ing the number of word senses of a target word in untagged data using a stability criterion, and then (2) grouping the instances of this target word into the estimated number of clusters according to the simi- larity of contexts of the instances. No sense-tagged data is used to help the clustering process. The goal of task 5 (Chinese Word Sense Disam- biguation) is to create a framework for the evaluation of word sense disambiguation in Chinese-English machine translation systems. Each participates of this task will be provided with sense tagged train- ing data and untagged test data for 40 Chinese pol- ysemous words. The ”sense tags” for the ambigu- ous Chinese target words are given in the form of their English translations. Here we used a semi- supervised classification algorithm (label propaga- tion algorithm) (Niu, et al., 2005) to address this Chinese word sense disambiguation problem. The lexical sample subtask of task 17 (English Word Sense Disambiguation) provides sense-tagged …

Content truncated to 500 words