Bias in Conversational Search: The Double-Edged Sword of the Personalized Knowledge Graph

TL;DR

This paper examines how personalized knowledge graphs in conversational AI can introduce and amplify biases, discussing their types, measurement, and mitigation strategies to improve user satisfaction.

Contribution

It provides a comprehensive review of bias types in conversational search, focusing on PKGs, and proposes strategies for bias mitigation and evaluation methods.

Findings

Identifies various biases associated with PKGs in conversational AI.

Reviews existing bias definitions and measurement techniques.

Suggests strategies for bias reduction and improving user satisfaction.

Abstract

Conversational AI systems are being used in personal devices, providing users with highly personalized content. Personalized knowledge graphs (PKGs) are one of the recently proposed methods to store users' information in a structured form and tailor answers to their liking. Personalization, however, is prone to amplifying bias and contributing to the echo-chamber phenomenon. In this paper, we discuss different types of biases in conversational search systems, with the emphasis on the biases that are related to PKGs. We review existing definitions of bias in the literature: people bias, algorithm bias, and a combination of the two, and further propose different strategies for tackling these biases for conversational search systems. Finally, we discuss methods for measuring bias and evaluating user satisfaction.