Awards & Nominations
Visios has received the following awards and nominations. Way to go!
Hey! What Are You Looking At?
EnergyX
Summary
"EnergyX" is a website where the customer will find an easy and fast paced environment to access images, data and high quality content about space electromagnetic field phenomena from great resources, such as HEASARC, LAMBDA and CfA Supernova Group. More throughly, the main target by accessing the website, or rather, this visualization tool, is to get in touch to a platform of visual astrophysical content by using internally a web interface quite similar to HEASARC'S Xamin Web Interface, but based on an algorithm capable to connect and transform all the DBs related to these phenomena into real images. This is possible by setting up pixel format parameters to perform input info.
How I Addressed This Challenge
Through the algorithm depicted above and another as well devised to make predicitons on the images and photos taken by the satellite along its flight route, I can state the main target is "overachieved". We were meant to offer a visualization tool to let people watch and access images and data related to electromagnetic filed phenomena (gamma ray bursts, supernova, gravitational waves...) easily and quickly. The algorithm set up through our website is able to, on one side, transform the input data (from the institutional DBs and the upcoming info from the satellite) into real visual content. And as well, through the second ML algorithm, able to make predictions on whether the satellite will search out or not images leaning to our search criteria.
How I Developed This Project
I got focused on the web architecture/design implemented on the Xamin Web Interface, and then I realized how the network and the search criteria worked on the platform. Later, I just used the store or data files as a resource, and started to develop my own website with a similar interface where I could store that contect and information from the other two institutions. Then I realized, the current interface of course did not let people watch real images, so that I devised a ML algortihm inside the DOCTYPE HTML PUBLIC website code the AI script, just above the banner images section, in order to let it run along all the info provided above and executed below. At the moment, it does not work correctly due to some firewall protocols to sort out.
Later, for other projects, I had already devised another ML R script capable to make predictions on a binary statistics (Bernoulli) model, using t-SNE technique to shrink the dimensional data size, and then to calculate ROC, AUC and mistake (cost) function. Later, I focus on the correlation among the output data from the datasets and the functions in the training set, and proceed to make predictions for the test set.
At the moment, inside the website, it roughly works, but, due to the firewall problems, the results pop up in data format. It is just about fixing and establishing again the firewall configuration.
How I Used Space Agency Data in This Project
I searched out and reached out to very useful content at the adviced HEASARC and CADC websites, specially in the first one, which took me away to the LAMBDA archives and the Center for Astrophysics Supernova Group, founded by Harvard University and Smithsonian Observatory.
Project Demo
https://github.com/AMateos91/NASA-Space-Apps-Challenge-2020.git
In this GitHub repo of mine, everyone will be able to get access to all the work related to EnergyX.
Data & Resources
https://heasarc.gsfc.nasa.gov/FTP/
https://spdf.gsfc.nasa.gov/
https://heasarc.gsfc.nasa.gov/docs/xte/XTE.html
https://www.cfa.harvard.edu/supernova/
https://www1.cadc-ccda.hia-iha.nrc-cnrc.gc.ca/en/