Term Extraction for Micro-domain Ontology in Personalized Search

The objective of this project is to improve the quality of results returned by a search engine by incorporating the user’s prior search history. By adding contextual information about the user, we are able to better understand their needs and provide information tailored specifically for them. A unique, customized ontology is created to best model the user’s interests. The ontology is a signature of the user and can be used to properly disambiguate their query as well as quickly identify their domain of interest. Since the ontology is comprised of words and their relationships, by applying Natural Language Processing techniques along with machine learning methods that minimize the need for expensive prior knowledge, a semantic network is constructed. This ontology, which succinctly captures the user’s profile, is one of the core components in the target product for the partner organization. Our work provides a proof-of-concept for the target product.

Faculty Supervisor:

Anoop Sarkar

Student:

Partner:

Discipline:

Computer science

Sector:

University:

Simon Fraser University

Program:

Accelerate