WASH Analytics

Water and sanitation data analytics:

Introduction to the problem statement

Bringing analytics based on water and safely managed sanitation used in different countries and plotting how the countries are performing. Analyse and predict the mortality rate attributed to unsafe WASH services across the world.

Abstract

It is a fundamental and basic need to have pure drinking water and proper sanitation facilities in order to lead a healthy life. Although many people have gained access to basic sanitation services there are still many people especially in the rural areas lacking basic services. In the year 2016, water, sanitation and hygiene was responsible for 829 000 annual deaths from diarrhoea, and 1.9% of the global burden of disease. This makes this risk factor an important environmental contributor to ill health. Most diarrhoeal deaths in the world are caused by unsafe water, sanitation or hygiene. All together, improvements related to drinking-water, sanitation, hygiene, and water resource management could result in the reduction of almost 10% of the total burden of disease worldwide.

We are building a model that plots the performance of different countries in a graphical manner by training the dataset on the various water and sanitation facilities across the countries. We will be using all the existing WASH and mortality rate data to train our model and plot prediction graphs for mortality rate based on coverage of different water service levels.

Approach

The first part analysis will be done on the water sanitation and mortality rate data. We will visualize the data using Machine Learning and will perform Exploratory Data Analysis on it and then will plot the graphs accordingly between countries or regions and basic water and sanitation services. In the UI part there will be options to search for countries or regions for which the user wants to visualize the data and based on the search criteria graphs will be plotted depicting the performance with respect to water and sanitation facilities. We will be predicting mortality rate with linear regression, decision tree regression and k-NN regression models and show the graphs in the UI.

Persona

• This application is intended for health care and social workers across the world to understand the water sanitation levels and mortality rates attributed to unsafe WASH services.
• This application also provides predictions about how the quality of WASH services should be imporoved inorder to cut down the mortality rates due to unsafe WASH services.

Dataset links (if any)

https://www.kaggle.com/aliyatyshkanbayeva/analysis-on-drinking-water-sanitation-and-hygiene

https://www.who.int/data/gho/data/indicators/indicator-details/GHO/mortality-rate-attributed-to-exposure-to-unsafe-wash-services-(per-100-000-population)-(sdg-3-9-2)

Team members

1. Sherin Thlakulathil Elias

2. Karan Pinakinbhai Jariwala

3. Parshv Nileshkumar Patel

Architecture Diagram

Technology Stack

• Dash
• Python
• ReactJS
• Plotly
• Flask
• Pandas
• Sklearn
• Google Colab
• AWS