@ University of North Carolina at Chapel Hill | School of Information and Library Science
Primary Researcher / Experimental Design + Usability Testing
Efficient Product Search Tools for Online Consumers with Little Product Knowledge: Clustering Versus Faceted Search
In this research, which led to my Master’s paper at UNC-Chapel Hill, I designed and conducted an experimental usability study by comparing three different shopping search interfaces (different types of facets/filters).
In addition to the types of shopping search interfaces, I recruited participants who have different domain knowledge backgrounds related with specific product items and explored the relationships between people’s domain knowledge (product knowledge) and their user experience in searching for specific products. Major findings included that the shopping search interfaces with the most specific types of facets (filters) provided the most efficient shopping search experience to users with little product knowledge.
Project Outcome: Master’s Paper, Efficient Product Search Tools for Online Consumers with Little Product Knowledge: Clustering Versus Faceted Search
Abstract: This study compares and contrasts different types of search tools—clustering search, rich-faceted search, and less-faceted search—in terms of their performance in online shopping product searches. In the experiment, two groups of people, a music-expert group and a computer-expert group, conducted searches for two different products (one for a novice search and the other for an expert search). Search performance was measured on three different shopping search websites—Clusty Shopping, PriceGrabber, and Yahoo Shopping. In addition, the participants’ perceptions of each website’s usefulness, directability/controllability, and their likelihood of future use are measured through their subjective ratings in questionnaires. The results of this research provide insight into how online consumers with different levels of product knowledge experience different types of search tools, with regard to search performance, perceived usefulness and controllability, and website preference.