You are invited to Saket Kumar's thesis defense.
Thesis Title: Word Sense Disambiguation
Thesis Advisor: Dr. Omar El Ariss
Date/Time: Tuesday, March 24, 2015, 11:15am
Location: Olmsted E310.
Abstract
Semantics or the meaning of a given sentence is an important step in communication and knowledge acquisition. Humans are good with understanding the meaning of a given text, but how do we do that? Is it simply looking up the definition of each word one at a time or is it more than that? Also, is it possible for a program to automate the process of language understanding?
Humans can infer meaning through the use of not only the definition of word, where one word might have various conflicting definitions, but also based on their experience and the text’s context and domain. Word Sense Disambiguation is a problem in natural language processing that automates the process of text interpretation. In other words, WSD is an algorithm that finds the most appropriate word meaning in a particular context. The importance of Word Sense Disambiguation (WSD) lies in the processing of large amount of data without the need of constant help and human intervention. It is crucial for many applications such as translation, summerization, information retrieval and many other natural language applications.
We introduce an unsupervised knowledge based approach for word sense disambiguation using a bee colony optimization algorithm. We also present several variations to our bee colony approach that improves the overall performance of the algorithm. The Results are compared with recent unsupervised approaches such as ant colony optimization, genetic algorithm, most frequent sense and simulated annealing.
Thesis Title: Word Sense Disambiguation
Thesis Advisor: Dr. Omar El Ariss
Date/Time: Tuesday, March 24, 2015, 11:15am
Location: Olmsted E310.
Abstract
Semantics or the meaning of a given sentence is an important step in communication and knowledge acquisition. Humans are good with understanding the meaning of a given text, but how do we do that? Is it simply looking up the definition of each word one at a time or is it more than that? Also, is it possible for a program to automate the process of language understanding?
Humans can infer meaning through the use of not only the definition of word, where one word might have various conflicting definitions, but also based on their experience and the text’s context and domain. Word Sense Disambiguation is a problem in natural language processing that automates the process of text interpretation. In other words, WSD is an algorithm that finds the most appropriate word meaning in a particular context. The importance of Word Sense Disambiguation (WSD) lies in the processing of large amount of data without the need of constant help and human intervention. It is crucial for many applications such as translation, summerization, information retrieval and many other natural language applications.
We introduce an unsupervised knowledge based approach for word sense disambiguation using a bee colony optimization algorithm. We also present several variations to our bee colony approach that improves the overall performance of the algorithm. The Results are compared with recent unsupervised approaches such as ant colony optimization, genetic algorithm, most frequent sense and simulated annealing.
No comments:
Post a Comment