AI/ML intern
Computer Code Programming
Computer Programming/CodingThe project’s scope encompasses an end-to-end machine learning pipeline that combines K-Means clustering and an XGBoost classifier to predict airline passenger satisfaction. The project size is substantial, utilizing a dataset of approximately 129,880 passenger records with 22 distinct features. Notably, this project does not involve manual data labeling; it utilizes a pre-split, pre-labeled dataset. The specific data preparation tasks performed include programmatic label-encoding and one-hot-encoding of categorical variables , as well as algorithmically generating and appending new cluster labels. Quality measures strictly adhered to include systematically handling missing data , scaling numerical features with StandardScaler , utilizing Silhouette Scores for optimal K selection , and rigorously evaluating the final model using Accuracy, F1-Score, and ROC-AUC metrics on a held-out test set.