Pump it Up: Data Mining the Water Table
The problem statement is part of DrivenData competition called "Pump it Up: Data Mining the Water Table".
Taarifa and Tanzanian Ministry of Water have provided data about water pumps installed in various sub villages in Tanzania for us to predict whether the water pump installed is Functional, Non-Functional or Needs Repair. The outcomes of processing this model will help the Tanzanian Ministry of Water to improve their maintenance operations of water pumps and ensure clean water is available across the sub villages.
This Repository contains:
- 3 data files (.csv) namely: ‘training_set_values.csv’, ‘training_set_labels.csv’ and ‘testing_set_values.csv’.
- 2 processed data (.csv) files: These are the data files obtained after preprocessing the data. These are: pump_train_for_models (1).csv and water_table_test.csv.
- "PY files" folder contains the python codes for Preprocessing and the classifiers: Extreme Gradient Boosting, Random Forest and Gradient Boosting.