Visualization aids to support the consumer decision making process
Abstract
Many consumers today purchase products through online shopping websites. In the online shopping environment, consumers often suffer from information overload because they have limited cognitive resources to process the deluge of information, to evaluate multiple alternatives and attributes, and to reach an optimal choice. Using consumer reviews to make a purchase decision is one of the most challenging roadblocks. Since online consumer reviews tend to be text-heavy, deriving insights from collections of text reviews is very difficult for individual consumers. Visualization aids may help consumers organize and make sense of valuable information from consumer reviews, thereby increasing the quality of and satisfaction from purchase decisions. However, previous studies have not provided efficient methods to visually explore and manage consumer reviews. Therefore, the goal of this study is to develop and evaluate visualization aids that can help consumers explore and manage consumer reviews for their decision making. To achieve that goal, the author designed three series of studies. The first study examines whether consumers feel information overload while exploring many consumer reviews without specific visual help and how they overcome such a barrier. In the second study, the author develops and evaluates a visualization called ThemeReview, which enables consumers to explore reviews based on their characteristics such as themes, review lengths, helpfulness, and rating score. The third study proposes to use a simple mark-up technique to capture product features and opinions while reviewers are writing their reviews so that visualizations can use the feature data in a high quality. Throughout the three studies, the author found that consumers tend to rely heavily on consumer ratings (e.g., star ratings) due to the current user interface of online shopping websites (e.g., Amazon.com). A review visualization embedded in online shopping websites, such as ThemeReview, can help consumers explore more reviews and make more confident decisions based on data. The author also believes that the review summarization data can be improved even more by using a crowd's efforts. This research contributes to the expansion of visualization opportunities in a casual visual analytics and consumer decision making domain.
Degree
Ph.D.
Advisors
Yi, Purdue University.
Subject Area
Industrial engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.