Theses and Dissertations


Issuing Body

Mississippi State University


Bethel, Cindy L.

Committee Member

Swan, J. Edward, II

Committee Member

Torri, Stephen

Committee Member

Bhowmik, Tanmay

Date of Degree


Document Type

Dissertation - Open Access


Computer Science

Degree Name

Doctor of Philosophy (Ph.D)


James Worth Bagley College of Engineering


Department of Computer Science and Engineering


Several studies have applied recommender technologies to support requirements engineering activities. As in other application areas of recommender systems (RS), many studies have focused on the algorithms’ prediction accuracy, while there have been limited discussions around users’ interactions with the systems. Since recommender systems are designed to aid users in information retrieval, they should be assessed not just as recommendation algorithms but also from the users’ perspective. In contrast to accuracy measures, user-related issues can only be effectively investigated via empirical studies involving real users. Furthermore, researchers are becoming increasingly aware that the effectiveness of the systems goes beyond recommendation accuracy, as many factors can be relevant to their adoption besides accuracy. To better understand recommender systems in RE, it has become necessary to explore them from users’ perspectives. Consequently, this research evaluates a feature recommender system from users’ perspectives adopting the “Recommender systems’ Quality of user experience” (ResQue) model - a user-centered evaluation model from the RS field. This was done by designing a content-based feature recommender system and then assessing it from the users’ view point. A between-subjects user study was conducted involving two groups of participants, an experimental and a control group. The experimental group interacted with the feature recommender system while developing a list of software requirements for a software product (an antivirus software). In contrast, the control group performed the same task without receiving support from the recommender. After completing the task, both groups completed a post-task evaluation questionnaire, including questions about their experiences and opinions about the task they completed. In addition, participants in the experimental group rated their perceptions of various aspects of the recommender; question items were adapted from the ResQue questionnaire. Users’ subjective evaluation of the recommender was investigated using the ResQue constructs - perceived system qualities, user beliefs, user attitudes, and behavioral intentions. Additionally, the impact of recommendations on the requirements elicitation process was assessed in terms of the process and outcome level measures. Possible qualitative differences were also examined. Users' preferences were identified, and possible HCI issues requiring attention in recommender systems used in RE are discussed.