GENDER PREFERENCES IN CRYPTOCURRENCY SYSTEMS: SENTIMENT ANALYSIS AND PREDICTIVE MODELLING IN THE KINGDOM OF BAHRAIN

This study investigates gender-based differences in cryptocurrency preferences using sentiment analysis and predictive modelling within the financial technology ecosystem of the Kingdom of Bahrain. As cryptocurrencies continue to reshape digital financial markets, understanding demographic and behavioural factors influencing cryptocurrency adoption has become increasingly important. This research employs a quantitative survey-based approach involving 620 respondents to examine demographic characteristics, digital engagement patterns, and cryptocurrency investment behaviours. Descriptive statistical analysis was conducted to evaluate respondent characteristics, while machine learning techniques, including Logistic Regression and Random Forest classifiers, were implemented to predict cryptocurrency preferences. The findings reveal notable gender disparities in cryptocurrency engagement and investment preferences. Male participants demonstrated higher participation levels and stronger predictive classification accuracy compared to female respondents. The Random Forest model achieved the highest predictive performance with an accuracy of 87.4% and an F1-score of 0.85, significantly outperforming the Logistic Regression model. Feature importance analysis further identified gender as a significant predictor of cryptocurrency preference, suggesting distinct behavioural patterns between male and female investors. The study contributes to the growing literature on financial technology adoption by providing empirical insights into gender-specific cryptocurrency investment behaviour in emerging digital economies. The findings offer practical implications for fintech developers, financial institutions, and policymakers seeking to design more inclusive and user-centric cryptocurrency platforms.