Duluth: Word Sense Discrimination in the Service of Lexicography

Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015), pages 438–442, Denver, Colorado, June 4-5, 2015. c⃝2015 Association for Computational Linguistics Duluth: Word Sense Discrimination in the Service of Lexicography Ted Pedersen Department of Computer Science University of Minnesota Duluth, MN, 55812, USA tpederse@d.umn.edu Abstract This paper describes the Duluth systems that participated in Task 15 of SemEval 2015. The goal of the task was to automatically construct dictionary entries (via a series of three sub- tasks). Our systems participated in subtask 2, which involved automatically clustering the contexts in which a target word occurs into its different senses. Our results are consis- tent with previous word sense induction and discrimination findings, where it proves diffi- cult to beat a baseline algorithm that assigns all instances of a target word to a single sense. However, our method of predicting the num- ber of senses automatically fared quite well. 1 Introduction A Corpus Pattern Analysis (CPA) dictionary en- try building task (SemEval 2015 Task 15) included three subtasks, the combination of which creates a dictionary entry based on CPA (Hanks, 2013). The Duluth systems participated in the second subtask, which sought to cluster the contexts in which tar- get words occur based on their underlying sense or meaning. Note that for this task all of the target words are verbs. This is unusual for a word sense shared task, since nouns are much more commonly studied. The task input includes two sets of words : the Microcheck includes 8 target verbs, where the num- ber of senses for each are given to task participants, while the Wingspread includes 20 target verbs where the number of senses are withheld. Both sets of tar- get verbs and their frequencies are shown in Tables 3.2 and 3.2. The CPA method is based on finding patterns of use in corpora, and definitions of word senses re- fer explicitly to these patterns. For example, the verb totter has three senses, where a person (sense 1), building (sense 2), or institution (sense 3) may be what totters. The verb undertake has two senses, where a person or institution embarks on an activity (sense 1) or promises to do so (sense 2). There is certainly a role for syntactic information in defining such senses – direct and indirect objects are clearly important, and chunking would in gen- eral be quite useful. It also seems that incorporating semantic features, for example, those based on se- lectional restrictions or constraints, might be fruit- ful. In fact, subtask 1 focuses on shallow parsing and is said to be similar to semantic role labeling. Given different syntactic and semantic features dis- covered in subtask 1, it would be possible to pursue subtask 2 using a more rule based approach. However, the Duluth systems do not explicitly ac- count for syntax or semantics and do not try to iden- tify these kinds of patterns. While we believe such approaches are extremely useful, we are primarily interested in exploring the limits of …

Content truncated to 500 words