Predict functionality of Tanzanian water pumps.
Here are my attempts at the competition 'Pump It Up: Data Mining the Water Table'. In the first notebook I explored the data set and made it suitable for running models on. The first model I attempted was a Linear Support Vector Classifier which achieved a score of 0.7318 (the scale is from 0 to 1 with 1 being a perfect score). Then I used a more Gradient Boosting Classifier which gave the best score out of all the models I used: 0.8073. Finally, I used XGBoost which produced a score of 0.6936.
The data used and more information can be obtained here: https://www.drivendata.org/competitions/7/