Dal Alert!

Receive alerts from Dalhousie by text message.

X

MACSc Project Presentation - Using Efficient Data Structure to Speed Up Word Similarity Queries on WordNet

Who: Saurabh Singh

Title: Using Efficient Data Structure to Speed Up Word Similarity Queries on WordNet

Examining Committee:

Dr. Norbert Zeh - Faculty of Computer Science (Supervisor)
Dr. Vlado Keselj - Faculty of Computer Science (Co-Supervisor)
Dr. Abidalrahman Moh'd - Faculty of Computer Science (Reader)

Abstract:

The quantification of semantic similarity has many applications in natural language processing, and many different measures are proposed using Wordnet as their central resource. Wordnet is a lexical database for the English language. It groups different English words under different parts of speech into sets of words with very similar meaning

called synsets. All synsets are connected to other synsets using semantic relations. Wordnet is primarily used in automatic text analysis and artificial intelligence applications. We implemented an algorithm for fast evaluation of semantic similarity based on WordNet using efficient data structures. We discuss the motivation and implementation of these algorithms along with the data structures they use and present evaluation results compared with the well-known NLTK toolkit.

Time

Location

Room 430, Goldberg Computer Science Building