Word Sense Disambiguation Using WordNet Semantic Knowledge

Word Sense Disambiguation (WSD) has been an important and difficult problem in Natural Language Processing (NLP) for years. This paper proposes a novel WSD method which expands the knowledge for senses of ambiguous word through semantic knowledge in WordN

  • PDF / 540,457 Bytes
  • 10 Pages / 439.37 x 666.142 pts Page_size
  • 52 Downloads / 231 Views

DOWNLOAD

REPORT


Abstract Word Sense Disambiguation (WSD) has been an important and difficult problem in Natural Language Processing (NLP) for years. This paper proposes a novel WSD method which expands the knowledge for senses of ambiguous word through semantic knowledge in WordNet. First, selecting feature words through syntactic parsing. Second, expanding the knowledge for the ambiguous word senses through glosses and structured semantic relations in WordNet. Third, computing the semantic relevancy between ambiguous word and context and achieving the purpose of WSD by semantic network in WordNet. Lastly, adopting the Senseval-3 all words data sets as the test set to evaluate our approach. Through a detailed experimental evaluation, the result shows that our approach achieves improvements over some classical methods. Keywords Word sense disambiguation

 Syntactic parsing  Semantic relevancy

1 Introduction In Natural Language Processing (NLP), it is common that a word has multiple meanings. Word Sense Disambiguation (WSD) is to exploit an ambiguous word and determine which sense of the word should be assigned in the given context. N. Gao (&)  W. Zuo (&)  Y. Dai  W. Lv College of Computer Science and Technology, Jilin University, Changchun, China e-mail: [email protected] W. Zuo e-mail: [email protected] Y. Dai e-mail: [email protected] W. Lv e-mail: [email protected]

Z. Wen and T. Li (eds.), Knowledge Engineering and Management, Advances in Intelligent Systems and Computing 278, DOI: 10.1007/978-3-642-54930-4_15,  Springer-Verlag Berlin Heidelberg 2014

147

148

N. Gao et al.

WSD is a long-standing problem in NLP, which has broad impact on many important NLP applications, such as machine translation, information retrieval, and question answering. The human beings can distinguish ambiguous word through other words in context. To simulate the process of human thinking, we can extract useful information from the given context and then use it to achieve the purpose of WSD. Therefore, in order to determine which sense of ambiguous word should be adopted in the given context, we need two resources:  the context in which the ambiguous word has been used, in general, can be represented by feature words set of it. There are two commonly used methods to select feature words: windowbased methods and dependency-based methods; ` some kind of knowledge, related to senses of ambiguous word, constitutes the main basis for comparison with the context. A variety of machine-readable dictionaries and ontologies (e.g., WordNet [1]) can be used as knowledge sources of word sense’s knowledge. Due to the advantages of WordNet (we will describe it later), we choose it as the knowledge source of our method. Also, we need a good method to compute the semantic relevancy [2] between ambiguous word and context. In recent years knowledge-based WSD approaches have become quite popular. This paper proposes the SKW-WSD (SKW means semantic knowledge in the WordNet) method. This method utilizes nouns in the gloss and semantic relationships to expend the k

Data Loading...