PRE-LUTE

Helping predict, prevent and educate users about Pollution

Comment

Inspiration

Around 40% of the lakes in America are too polluted for aquatic life, swimming or fishing.Although children make up 10% of the world’s population, over 40% of the global burden of disease falls on them. Environmental factors contribute to more than 3 million children under age five dying every year. Pollution kills over 1 million seabirds and 100 million mammals annually. Recycling and composting alone have avoided 85 million tons of waste to be dumped in 2010. Currently in the world there are over 500 million cars, by 2030 the number will rise to 1 billion, therefore doubling pollution levels. High traffic roads possess more concentrated levels of air pollution therefore people living close to these areas have an increased risk of heart disease, cancer, asthma and bronchitis. Inhaling Air pollution takes away at least 1-2 years of a typical human life. 25% deaths in India and 65% of the deaths in Asia are resultant of air pollution. Over 80 billion aluminium cans are used every year around the world. If you throw away aluminium cans, they can stay in that can form for up to 500 years or more. People aren’t recycling as much as they should, as a result the rainforests are be cut down by approximately 100 acres per minute On top of this, I being near the Great Lakes and Neeral being in the Bay area, we have both seen not only tremendous amounts of air pollution, but marine pollution as well as pollution in the great freshwater lakes around us. As a result, this inspired us to create this project.

What it does

For the react native app, it connects with the Website Neeral made in order to create a comprehensive solution to this problem.

There are five main sections in the react native app:

The first section is an area where users can collaborate by creating posts in order to reach out to others to meet up and organize events in order to reduce pollution. One example of this could be a passionate environmentalist who is organizing a beach trash pick up and wishes to bring along more people. With the help of this feature, more people would be able to learn about this and participate.

The second section is a petitions section where users have the ability to support local groups or sign a petition in order to enforce change. These petitions include placing pressure on large corporations to reduce carbon emissions and so forth. This allows users to take action effectively.

The third section is the forecasts tab where the users are able to retrieve data regarding various data points in pollution. This includes the ability for the user to obtain heat maps regarding the amount of air quality, pollution and pollen in the air and retrieve recommended procedures for not only the general public but for special case scenarios using apis.

The fourth section is a tips and procedures tab for users to be able to respond to certain situations. They are able to consult this guide and find the situation that matches them in order to find the appropriate action to take. This helps the end user stay calm during situations as such happening in California with dangerously high levels of carbon.

The fifth section is an area where users are able to use Machine Learning in order to figure out whether where they are is in a place of trouble. In many instances, not many know exactly where they are especially when travelling or going somewhere unknown. With the help of Machine Learning, the user is able to place certain information regarding their surroundings and the Algorithm is able to decide whether they are in trouble. The algorithm has 90% accuracy and is quite efficient.

How I built it

For the react native part of the application, I will break it down section by section.

For the first section, I simply used Firebase as a backend which allowed a simple, easy and fast way of retrieving and pushing data to the cloud storage. This allowed me to spend time on other features, and due to my ever growing experience with firebase, this did not take too much time. I simply added a form which pushed data to firebase and when you go to the home page it refreshes and see that the cloud was updated in real time

For the second section, I used native base in order to create my UI and found an assortment of petitions which I then linked and added images from their website in order to create the petitions tab. I then used expo-web-browser, to deep link the website in opening safari to open the link within the app. For the third section, I used breezometer.com’s pollution api, air quality api, pollen api and heat map apis in order to create an assortment of data points, health recommendations and visual graphics to represent pollution in several ways. The apis also provided me information such as the most common pollutant and protocols for different age groups and people with certain conditions should follow. With this extensive api, there were many endpoints I wanted to add in, but not all were added due to lack of time.

For the fourth section, it is very much similar to the second section as it is an assortment of links, proofread and verified to be truthful sources, in order for the end user to have a procedure to go to for extreme emergencies. As we see horrible things happen, such as the wildfires in California, air quality becomes a serious concern for many and as a result these procedures help the user stay calm and knowledgeable.

For the fifth section, Neeral please write this one since you are the one who created it.

Challenges I ran into

API query bugs was a big issue in formatting back the query and how to map the data back into the UI. It took some time and made us run until the end but we were still able to complete our project and goals.

What's next for PRE-LUTE

We hope to use this in areas where there is commonly much suffering due to the extravagantly large amount of pollution, such as in Delhi where seeing is practically hard due to the amount of pollution. We hope to create a finished product and release it to the app and play stores respectively.

Built With

Share this project:

Updates

posted an update —

On behalf of Neeral, this is quick explanation of the ML he left me before we called it a night:

For the fifth section, I made a machine learning algorithm that detects if people are living in areas that have a high, medium, or low risk of air pollution. The machine learning algorithm uses regression to determine the best model and has a high accuracy rate.

Log in or sign up for Devpost to join the conversation.