i am a final-year student in ai and ds descipline.
my niche are ai, ml, dl...(under the hood machine learning).
A hybrid deep learning model combining CNN and LSTM layers to detect manipulated or deepfake audio recordings.
A customer churn analysis project using multiple models. XGBoost performed the best, with SHAP values and feature importance explaining the key drivers of churn.
This project predicts whether an online shopper will complete a purchase. It includes feature analysis, preprocessing, SMOTE balancing, model training, and evaluation using accuracy, reports, ROC curves, and AUC scores.
A project that explores customer happiness prediction by cleaning the data, removing weak features, testing multiple models, and visualizing results. The dataset lacked strong patterns, limiting prediction accuracy.