Publication Date

8-2023

Date of Final Oral Examination (Defense)

July 2023

Type of Culminating Activity

Dissertation

Degree Title

Doctor of Philosophy in Computing

Department

Computer Science

Major Advisor

Michael D. Ekstrand, Ph.D.

Advisor

Sole Pera, Ph.D.

Advisor

Edoardo Serra, Ph.D.

Advisor

Casey Kennington, Ph.D.

Abstract

Information access systems, such as search engines and recommender systems, often display results in ranked order based on their estimated relevance. The fairness of these rankings has received attention as an important evaluation criteria along with traditional metrics capturing constructs such as utility or accuracy. Fairness has many facets, including provider and consumer-side fairness at both group and individual levels. Research on provider-side group fairness involve concerns regarding measurement and optimization of fairness in ranking. Although there are several fair ranking metrics to measure provider-side group fairness based on various “sensitive attributes”, multiple open challenges still exist in this area to consider. Moreover, the fair ranking research mostly focuses on linear layouts when items are displayed in single-column list, often overlooking fairness issues in other layouts such as grid view.

In my dissertation, I work on the area of provider-side group fairness in ranking in information access systems. I seek to understand the fairness concepts and practical applications of existing fair ranking metrics and find ways to improve the metrics. My work wilaid researchers and practitioners in selecting fair ranking metrics by pointing out the strengths, limitations, applicability and reliability of the metrics. Moreover, I contribute to the advancement of fair ranking metrics by considering various ranking layout models and further contribute to provider-side group fairness optimization in ranking in widely-used but seldom-studied grid layout.

DOI

https://doi.org/10.18122/td.2127.boisestate

Download

Share

COinS