Electronic Theses and Dissertations

Identifier

663

Date

2012-07-20

Document Type

Thesis (Campus Access Only)

Degree Name

Master of Science

Major

Electrical and Computer Engr

Concentration

Computer Engineering

Committee Chair

Mohammed Yeasin

Committee Member

Aaron L Robinson

Committee Member

Bashir Morshed

Abstract

Significant progress has been made in applying text mining to named entity recognition, text classification, terminology extraction, relationship extraction and hypothesis generation. Businesses, government agencies, and researchers are gaining a competitive advantage through the use of text mining and content analytics from unstructured data. However, mining and classification of short-text with high accuracy remains a formidable challenge. To address this problem, a semantic-sensitive query expansion technique was developed and applied on a number of corpuses. The Query Expanded Latent Semantic Analysis (QE-LSA) model was developed from large short-text corpus with user defined, data driven and dynamic dictionary. Empirical analyses on two datasets obtained from different domains were conducted to illustrate the efficacy of QE-LSA. A number of experiments were conducted to understand the role of various tuning of parameters in maximizing the performance. The results suggest that this model performs robustly on both corpuses.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.

Download

Share

COinS