Leveraging Machine Learning to Analyze User Conversion in Mobile Pharmacy Apps Using Behavioral and Demographic Data

Abstrak

This study explores the use of machine learning techniques to predict user conversion in a mobile pharmacy app based on user behavior and demographic data. The analysis was conducted using two classification models: Logistic Regression and Random Forest. Key features such as time spent on the product page, adding items to the cart, and user demographics (age, gender, device type) were evaluated to determine their impact on conversion rates. Both models demonstrated strong performance, with the Logistic Regression model achieving an Area Under the Curve (AUC) of 0.88 and the Random Forest model achieving an AUC of 0.87. These results indicate that both models effectively distinguish between users who convert and those who do not, with Logistic Regression showing a slightly better overall performance. Feature importance analysis revealed that factors such as adding items to the cart and the time spent on the product page are the most significant predictors of conversion. Furthermore, demographic features like age group and device type also contributed to the model’s predictive power, although they had a smaller impact compared to user engagement features. The findings suggest that machine learning models, particularly Logistic Regression, can be utilized to predict user behavior and optimize user engagement strategies in mobile apps. The study also highlights the importance of user engagement in driving conversions and the potential for targeted marketing based on demographic data. Future work should focus on hyperparameter tuning, exploring additional algorithms, and incorporating real-time data to further enhance model accuracy and adaptability.