Polaris has received the following awards and nominations. Way to go!
In recent years, we have witnessed many environmental disasters around the world. Among them, we can mention volcanic eruptions, earthquakes, tsunamis, hurricanes, landslides, and floods, in addition to global warming, caused by both human and natural actions, and which, unfortunately, has been taking many lives and damaging the local ecosystem.
To solve these problems, we created Polaris, a WEB application that uses machine learning and artificial intelligence technologies, using environmental, climatic, demographic, and socioeconomic data, collected in real-time from the satellites of several space agencies and available government data.
Thanks to this vast collection of data, Polaris can understand the reality of each region, preparing a specific analysis, according to the situation of each location. By taking environmental, demographic, and socioeconomic data as parameters, a predictable disaster platform can calculate the social and environmental damage caused by them.
For greater accuracy, Polaris learns over time the ecosystem patterns of all regions analyzed. From the moment that the collected data deviate from the standard, or fit into a pattern of environmental catastrophe, the system can identify the level of impact that can be issued and, alert as competent authorities, forwarding reports with the indicators used and the analysis of the data compared in time progression.
Despite its complex operation, Polaris presents a simple and intuitive interface for the user, making it easier to view and faster to make decisions.
The biggest inspiration was the desire to solve real pain in society, natural or provoked disasters. In 2019, in Brazil, we had a dam rupture in the city of Brumadinho - MG, which devastated the entire city, leaving a large portion of the population homeless, in addition to the hundreds of dead, and those who are still missing today. The pain of losing a loved one is generally felt by at least 10 relatives and/or friends, which increases the number of people affected (directly or indirectly) by a disaster by 10x.
With this in mind, we decided to create a system that, based on the collection of climatic, environmental, demographic, and socio-economic data, performs risk analysis, establishing standards for each region, informing the authorities when one of them deviates from its natural standards, and calculating the social impact of a potential catastrophe.
For this, we use tools such as machine learning and artificial intelligence, in addition to a dashboard, which works both WEB and Mobile, with a user-friendly interface, to facilitate the understanding of the analyzes.
For the code of these applications, we use tools such as React Native, Angular, MongoDB, JavaScript, and Python.
For our project, we use data from space agencies (NASA - Dust Storms, Air Quality, Wind and Temperature / JAXA - Rainfall) to carry out constant monitoring of ecosystem data from each region, to create a machine learning that consists of in identifying patterns and variations that could turn out to be a disaster. We also collect some open demographic and socioeconomic data (GOVERNMENTAL - Demographics, Socioeconomic data) to identify the level of impact that may be caused, uniting the environment, and the population that lives there.
Data from NASA: Air quality, Temperature, Wind and Dust storms;
Data from JAXA: Rainfall;
Governamental data: Demographics and Socioeconomic data.