dendar/Tanzania-s-Water-Quality-Analysis

# Tanzania's Water Quality Analysis and Predictive Models Report by Nadir Sarigul ## I. Overview and Goals This report contains a detailed explanation of the extraction, cleaning, analysis and generation of prediction models for the state of waterpoints in Tanzania. This report is based off of data collected from the Taarifa and the Tanzanian Ministry of Water as part of the "Pump it Up: Data Mining the Water Table" competition. The ultimate goal of this analysis is to predict which waterpoints will fail so that we can improve maintenance operations and ensure that clean, potable water is avaialble across Tanzania. To be able to understand the waterpoint conditions and functionality better, I divided this analysis in five questions: 1) Does geographical location affect functionality of the waterpoints? 2) Does the source of the water contribute for the functionality of the waterpoints? 3) Does the amount of water affect funtionality of the waterpoints? 4) Does payment type affect functionality of waterpoints? 5) Can we create a model that accuratly predicts the funtionality of a waterpoint in Tanzania? ## II. Data Understanding The data set that will be used in this analysis/model is comprised of 2 tables: - Training set labels: The dependent variable (status_group) for each of the rows in the training set values - Training set values: The independent variables for the training set This data set contains 59400 unique waterpoint entries containing information about the following features: * amount_tsh - Total static head (amount water available to waterpoint) * date_recorded - The date the row was entered * funder - Who funded the well * gps_height - Altitude of the well * installer - Organization that installed the well * longitude - GPS coordinate * latitude - GPS coordinate * wpt_name - Name of the waterpoint if there is one * num_private - * basin - Geographic water basin * subvillage - Geographic location * region - Geographic location * region_ …