End-to-end data science simulation for British Airways (Forage) — scraping 1,000 Skytrax customer reviews for NLP/sentiment analysis, and building two Random Forest classifiers on 49,281 booking records to predict booking completion (84.91% accuracy). Includes topic modeling, feature selection, and ROC/AUC evaluation.
python nlp machine-learning sentiment-analysis random-forest scikit-learn exploratory-data-analysis feature-selection web-scraping topic-modeling beautifulsoup predictive-modeling lda vader forage-simulation british-airways-forage
-
Updated
Apr 11, 2026 - Jupyter Notebook