Description

• Created a predictive model using Python and scikitlearn to predict customer interest in vehicle insurance based on health insurance purchases.
• Employed Machine Learning methodologies to analyze diverse datasets and extract valuable insights for cross-selling predictions.
• Ensured model transparency and interpretability by integrating LIME and SHAP for responsible AI implementation.
• Enhanced understanding of customer preferences and behaviors, leading to improved cross-selling strategies and business outcomes.
• Built a Streamlit application that creates an interactive dashboard to visualize model predictions, SHAP, and LIME explanations, enhancing user engagement and promoting transparency.
• Addressed data imbalance issues using Random Over-sampling Technique and evaluated its impact on model performance compared to non-balanced datasets.
• Secured 2nd position among all the submisions in the Responsible AI project hosted by HiCounsellor with an F1-score of 0.987.

Video of Streamlit Application

Certificate

Additional Information

The dataset is used from kaggle. The link to the notebook Link