Date of Award

Spring 1-1-2013

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Operations & Information Management

First Advisor

Kai R. Larsen

Second Advisor

Kenneth A. Kozar

Third Advisor

Jintae Lee

Fourth Advisor

Dirk S. Hovorka

Fifth Advisor

Nicolas Nicolov

Abstract

The advent of the Internet and the ever-increasing capacity of storage media have made it easy to store, deliver, and share enormous volumes of data, leading to a proliferation of information on the Web, in online libraries, on news wires, and almost everywhere in our daily lives. Since our ability to process and absorb this information remains relatively constant, there is an imperative demand for novel tools to help explore, extract, and understand this information. Information extraction and text mining are two research endeavors that seek to extract structured information and discover knowledge patterns from unstructured text data. Based on state-of-the-art information extraction and text mining techniques, this dissertation presents three essays that address the information proliferation in both academia and industry. Specifically, the first two essays focus on extracting constructs, theoretical models, and theory-specific citation patterns for the behavioral sciences, and the last essay aims to build a high-quality recommendation engine for the movie industry by combining textual and numerical information to create personalized output. The evaluation results for the system performance represent a promising opportunity to apply information extraction and text mining to the business domain.

Share

COinS