Skip to Main Content
Frequently Asked Questions
Submit an ETD
Global Search Box
Need Help?
Keyword Search
Participating Institutions
Advanced Search
School Logo
Files
File List
16192.pdf (4.15 MB)
ETD Abstract Container
Abstract Header
Essays in Business Analytics
Author Info
Mai, Feng
Permalink:
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1439295906
Abstract Details
Year and Degree
2015, PhD, University of Cincinnati, Business: Business Administration.
Abstract
The availability of structured and unstructured data, along with recent advancements in machine learning methods and tools, pose both challenges and opportunities for businesses. The three essays in this dissertation address important aspects of business such as marketing and operations using emerging business analytics methods. The essays are devoted to two topics in analytics: advances in unsupervised learning methods and analytics of unstructured, textual data. In Essay 1 we develop a business intelligence framework and advance market structure analysis by combining computational linguistics, machine learning, and relevant marketing theories to reveal consumer insights from free-form product reviews. Our text analytics method is able to create a hierarchy for product attributes, discover consumer sentiments, and construct market structure perceptual maps. In Essay 2, we use deep learning and evolutionary clustering to study the dynamics of market segmentation. We adopt the skip-gram model to learn computable, vectorized representation of product attributes. In addition, the evolutionary clustering model integrates a measure of temporal smoothness into the overall measure of clustering quality, and thus can be used as a method to study market structures over time. In Essay 3, we apply expectation-maximization (EM), a widely used method in statistical inference, to solve a discrete optimization problem that has many applications in operations management. We frame the optimization problem as a semi-supervised learning problem and develop a heuristic to solve a capacitated clustering problem and its stochastic variant.
Committee
Michael Fry, Ph.D. (Committee Chair)
Jeffrey Ohlmann, Ph.D. (Committee Member)
Hsiang-Li Chiang, Ph.D. (Committee Member)
David Curry, Ph.D. (Committee Member)
Subject Headings
Business Administration
Recommended Citations
Refworks
EndNote
RIS
Mendeley
Citations
Mai, F. (2015).
Essays in Business Analytics
[Doctoral dissertation, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1439295906
APA Style (7th edition)
Mai, Feng.
Essays in Business Analytics.
2015. University of Cincinnati, Doctoral dissertation.
OhioLINK Electronic Theses and Dissertations Center
, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1439295906.
MLA Style (8th edition)
Mai, Feng. "Essays in Business Analytics." Doctoral dissertation, University of Cincinnati, 2015. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1439295906
Chicago Manual of Style (17th edition)
Abstract Footer
Document number:
ucin1439295906
Download Count:
9,500
Copyright Info
© 2015, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.