Assessing the trust in male and female advisors during recommendation processes.

Abstract

Gender bias remains a pervasive issue in contemporary society, impacting various domains, including recommendation processes where individuals seek advice from advisors. The gender of these advisors significantly influences the perceived credibility and trustworthiness of the advice. This thesis aims to explore the dynamics of gender bias in advisory roles by examining preferences and trust levels towards male and female advisors across different domains. Specifically, we investigate whether individuals exhibit a noticeable preference for advisors of a particular gender and the factors contributing to these preferences. To address these questions, we analyze data from online platforms such as Quora, X, and StackOverflow, where users frequently seek advice. We extract data from these platforms and perform gender prediction using machine learning and deep learning techniques. Additionally, we employ feature extraction methods such as TF-IDF, Word2Vec, and GloVe to analyze the text data. By applying sentiment analysis, we uncover patterns of gender dominance and trust in the responses. Our research sheds light on the current state of gender bias in advisory contexts and provides insights for fostering more equitable and inclusive environments in professional and educational settings. The findings of this study contribute to understanding the underlying dynamics of gender bias and offer implications for addressing these biases effectively.