Water Quality Predictor

Inspiration

Safe and drinkable water is a necessity for the society to be healthy. Improved water quality, water management system can help the society grow economically and also health wise.

Poor sanitation and water contamination can lead to transmission of diseases like cholera, diarrhoea, and typhoid. We can avoid exposure to this diseases but understanding the water quality and by improving it.

This is what inspired us to build a model to predict the water quality.

What it does

This project predicts the water quality by using the Random Forest classifier. A webpage is created where the feature data should be entered by user.

How we built it

We used the water quality data set from Kaggle and using a Random forest classifier we trained the data . For training the model, we used 30% of water quality dataset. And the 70% of dataset was used to test the model. Initially we calculated the accuracy of following algorithms : Logistic Regression, K Neighbors Classifier, Decision Tree Classifier, GaussianNB, NearestCentroid, RandomForestClassifier, GradientBoostingClassifier and PassiveAggressiveClassifier.

Challenges we ran into

As we have no prior experience in machine learning, web data

Accomplishments that we're proud of

Even though our project is not completed, this has been a really great learning experience. We were able to build a model successfully, compare different models, and successfully use Flask to integrate with the machine learning model.

What we learned

What's next for Water Quality Predictor

As we were not able to successfully complete the project, our next step would be to complete it the project. We are getting error in the prediction part.