High-Level Project Summary
In this concept, two case studies are being explored to make it easier for people to use machine learning technology in minimizing environmental injustice events. Water pollution and climate change are two examples of environmental injustice. In this study, data on water quality and air temperature are used as case studies. Our proposal concept attempts to predict an index that shows whether it is safe for people to drink water, as in the first case study, or if it is safe for people to walk out on a given day, as in the second case study.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
In this concept, two case studies are being explored to make it easier for people to use machine learning technology in minimising environmental injustice events. Water pollution and climate change are two examples of environmental injustice. In this study, data on water quality and air temperature are used as case studies.
Our proposal concept attempts to predict an index that shows whether it is safe for people to drink water, as in the first case study, or if it is safe for people to walk out on a given day, as in the second case study.
In the first case study, if the output of our system is 1, it means the water is safe to drink; 0 implies the water is not safe to drink and must be treated at a water treatment plant. The coefficients are created via classification learning of the nine distinct water parameters. We tested our model's performance using a variety of machine-learning techniques, including Random Forest, decision tree, KNN, SVM, logistic regression, and Naive Bayes.
We utilized a water quality dataset with nine parameters, and then used 60% of that data to train the model and 40% to test the data. The findings demonstrate that Random Forest and Decision Tree beats the other methods because they have an accuracy of 100%. This case study may be extended to incorporate hardware, such as water sensors installed inside lakes to transfer data to the cloud, where our system is hosted for analysis or even can be implemented on the edge.
In the second case study, our system generates an indicator that displays the temperature difference from the baseline temperature based on learning the temperature's behavior throughout the period of the year. More climate change coefficients can be used in the future to demonstrate the influence of climate change on temperature. Citizens may rely on such a system to determine whether or not it is safe to go out that day. This technology is unique in that some countries lack a system that can provide an accurate temperature of the day towards climate change.
Unfortunately, this system was unable to consider climate change parameters because this data is not available on the internet, and studying these parameters would require a significant amount of time, which we do not have at the moment; however, a proof-of-concept of the idea is simulated using NASA air temperature data. In this case study, a profit model is used, with 60% of the dataset used as training data and 40% as testing data. As previously stated, this case study may be extended by incorporating climate change parameters into the learning process in order to get a more accurate air temperature index.
Finally, we developed two case studies in order to alleviate human suffering as a result of environmental injustice. The first case study generates an index that indicates whether or not the transferred water is safe to drink. The second case study aims to create an index that indicates whether or not it is safe for humans to go outside. This concept could be expanded to include other solutions to environmental injustice.
Space Agency Data
The temp. data has been collected by NASA and used in our model to predict the temp.
https://data.giss.nasa.gov/gistemp/tabledata_v4/GLB.Ts+dSST.txt
The water Dataset has been collected by one of the members using an external source.
https://data.world/
Hackathon Journey
We are very excited to be involved and part of this hackathon. our team has enjoyed the challenges and after meeting virtually we decided to narrow down the areas for the challenge and then we focused on the water quality issue and the temperature prediction as they are the most difficult issues that our world is facing. We loved the collaboration and work as a team to solve these issues.
Each one of the team focused on one thing, one of use worked on the data, the other one worked on the visualisation and the other one on the coding and ML algorithms.
References
https://data.giss.nasa.gov/gistemp/tabledata_v4/GLB.Ts+dSST.txt
Tags
#waterquality #Tempreture #BetterWorld #BetterFeature
Global Judging
This project has been submitted for consideration during the Judging process.