Downloads: 115

Research Paper | Computer Science & Engineering | India | Volume 3 Issue 6, June 2014

An Optimized Combinatorial Approach of Learning Algorithm for Word Sense Disambiguation

Neetu Sharma ^[6] | S. Niranjan ^[3]

Abstract: Word sense disambiguation is the process to find best sense of ambiguous word from the existing senses to remove the ambiguity. This thesis work is an attempt to optimize the word sense disambiguation method. Most commonly supervised machine learning algorithms were used to solve this problem and improve the performance. Some attempts were made to use unsupervised machine learning algorithms also like K-means clustering algorithm. In this research work supervised learning algorithm Nave Bayesian is combined with the unsupervised learning algorithm K-means Clustering and the performance is enhanced in getting best sense of ambiguous word. C# is used to create interface for getting input in the form of sentence containing ambiguous word and displaying the output as a best sense for that ambiguous word. SQL 2008 is used a s database to store the sentences entered and their corresponding meanings. WORDNET as a database for extracting senses of ambiguous word is used. Performance is evaluated on the basis of scores of precision; recall and F-score that how well this optimized algorithm works now to improve the accuracy.

Keywords: Nave Bayesian Algorithm, K-Means Clustering, WORDNET,

Edition: Volume 3 Issue 6, June 2014,

Pages: 2054 - 2061

Similar Articles with Keyword 'WORDNET'