IBM Applied Data Science Capstone
A comprehensive data science project predicting Falcon 9 first-stage landing success. Involves Web Scraping, Data Wrangling, EDA with SQL & Visualization, Interactive Dashboards (Dash), and Machine Learning Classification.
Proof points
- 83% Accuracy
- Interactive Dashboard
- Full DS Pipeline
Technologies
Python, Pandas, Scikit-learn, Dash, Folium, SQL
Links
Problem
SpaceX advertises significant savings due to rocket reusability. Competitors need to estimate launch costs by predicting if the first stage will land successfully.
Approach
Collected data via SpaceX API and web scraping. Cleaned and analyzed data using SQL and Pandas. Built interactive maps (Folium) and dashboards (Dash) for exploration. Trained and tuned classification models (Logistic Regression, SVM, KNN, Decision Tree) to predict landing outcomes.
Results
Achieved ~83.33% accuracy across all models. Identified that launch site, orbit type, and payload mass are key predictors of landing success. Delivered a full-stack data science solution from raw data to predictive insights.