Date of Award

2013

Publisher

North Dakota State University

Document Type

Thesis

Degree Awarded

Master of Science (MS)

Program

Computer Science

Department

Computer Science

College

Engineering

Faculty Advisor

Jin, Wei

Description

One major piece of information available on the web is reviews about various products that are written by users. Some commercial websites provide additional information about the products along with the reviews. However, in the opinion mining research field, most existing methods have ignored this additional valuable information, thus influencing the accuracy of the mining results and the interpretation of various aspects related to the products. In this thesis, we consider the reviews obtained from Epinions.com related to cameras, and we propose ReviewMiner, an unsupervised method of automatically identifying useful aspects of a product and estimating the corresponding ratings for each aspect from the review texts. The method explores various linguistic patterns to extract potential aspects and context-dependent opinion phrases, and the technique employs a series of heuristic strategies and pruning techniques. Experimental results have demonstrated the effectiveness of the proposed techniques and shown their advantages over comparative baselines.

Rights

NDSU Policy 190.6.2

Rights Link

https://www.ndsu.edu/fileadmin/policy/190.pdf

Handle Identifier

https://hdl.handle.net/10365/27183

Download

Share

COinS