kyawsawhtoon/Project-Tanzania-Water-Wells
Domain:
environment and energyRecord type:
project
Creator:
kya
Host:
The goal of this project is to construct a classification model that can be used to predict the condition of water wells in Tanzania in terms of 3 statuses: 'functional', 'functional but needs repair' and 'non functional'
# Repository Files
The followings are the file you can find in this repository and their descriptions.
Module 3 - Final Project (Pump it Up).ipynb - Module 3 Final Project's Jupyter Notebook
presentation.pdf - High-level presentation of methodology and recommendations for non-technical stakeholders
README.md - Descriptions of contents of the repository
TrainingSetValues.csv - The independent variables for the training set
TrainingSetLabels.csv - The dependent variable (status_group) for each of the rows in Training set values
TestSetValues.csv - The independent variables that need predictions
SubmissionFormat_Complete.csv - The predictions generated by the final model
Confusion Matrix.jpg - The confusion matrix chart of the final model
Performance Metrics Comparison.jpg - The horizontal bar chart that shows the performance scores of various models
Feature Importance.jpg - The horizontal bar chart that shows the importance of each independent variable
Top 5 Feature Importance.jpg - The horizontal bar chart that shows the importance of top 5 independent variable
# Project Motivation
The goal of this project is to construct a classification model that can be used to predict the condition of water wells in Tanzania. This model will classify the condition of water wells into 3 categories namely -
functional - the waterpoint is operational and there are no repairs needed
functional needs repair - the waterpoint is operational, but needs repairs
non functional - the waterpoint is not operational
# Project Process
My approach for this project is to follow the OSEMN framework. The steps of this framework is as follow -
Obtain the data
Scrub the data
Explore the data
Model the data
Interpret the data
# Data Source
To construct this model, I will use the data from Taarifa and the Tanzanian Ministry of Water. The original data is originally divided into 3 csv files -
Test set values: The independent variables that need predictions
Training set labels: …