We saw Esri's GIS platform and realized that it is a very useful visualization tool for a variety of applications, especially for visualizing data and to facilitate better machine learning prediction models.

What it does

Utilizing modern machine learning (GANs) and GIS, we can outperform existing site selection models for solar energy farms.

Additionally we can same methods to detect and analyze certain areas of the country that are posed with a larger risk of getting defrauded by investment advisors.

How we built it

For the data analysis portion where we scraped, extracted, cleaned, and analyzed from various sources we made extensive use of the python pandas framework. We acquired data from FINRA, arcGIS, US Census, OpenPV.

We implemented and trained our machine learning models in tensorflow and used matplotlib for analysis.

Our frontend access was done through the use of javascript, nodejs, and the arcGIS visualization api.

What we learned

GIS is a very underdeveloped tool when using location data in machine learning models.

What's next for Site Selection & Financial Analysis using GANs & GIS Data

Publishing a paper and writing a research report.

Share this project: