Amohelang Monaleli

I've developed an end-to-end machine learning system that processes raw drug-review data and classifies reviews into medical condition categories.

I explored, cleaned, and prepared data through missing-value handling, outlier treatment, statistical analysis, feature engineering, and NLP techniques including text cleaning, stop-word removal, stemming, lemmatisation, and TF-IDF extraction.

I compared Naïve Bayes, Decision Trees, and Logistic Regression, then trained, evaluated, and tuned the selected model using accuracy, precision, recall, F1-score, and Grid Search.

I built a functional Streamlit dashboard for real-time predictions, pattern analysis, and recommendations, with SQLite integration for storing user inputs, predictions, and recommendations.