AI training via supervised ML model development on customer churn and tweet sentiment datasets
Developed AI/ML models using supervised learning workflows, including model training, evaluation, and iterative experimentation. Built pipelines that preprocess and transform input data into features suitable for classification tasks. Worked with machine learning libraries in notebook environments to produce and validate predictive models. • Customer churn binary classification using a Random Forest model with reported accuracy • Sentiment classification pipeline using text preprocessing and TF-IDF vectorization • Trained and compared Naive Bayes and Logistic Regression models for NLP sentiment • Used scikit-learn with pandas/NumPy and evaluation metrics (accuracy, precision, recall, F1)
2024