Inspiration

We are inspired by the practical applications of this project in both the field of data science as well as its potential positive impact on both a local and global scale.

What it does

We created a program that establishes connections between stock brokers and investment advisors in a database according to location and the amount of time associated/ working with each other. The overall purpose of this project is to predict how corrupt an investment advisor is according to how "corrupt" their contacts are, which is determined by the number of contact "mentions" in a corpus of finance-related court cases. In other words, consider it a search engine for the corruption heatmap of a person based on the corruption level of their contacts

How we built it

Much of this project lives in python. We imported the data needed from https://www.courtlistener.com/api/rest/v3/ and stored the data into python dictionaries. From there, we wrote programs with the data that searched for people with "matching" qualities, such as with location.

Challenges we ran into

As of now, our project works well in python script. One big challenge comes in publishing the python script to a more easily accessible public source like the web. Our attempts to incorporate the graph and data processing into the webpage have all sadly failed. Filtering court data has also been challenging due to its sheer length.

Accomplishments that we're proud of

Introduction to programming at our school (UVA) is taught in python, so we learned the basics of python dictionaries, lists, strings, etc. in school. At this hackathon, we definitely pushed the limits of what we knew to create something very useful and solve a complex problem. We are also proud of how we were able to obtain a good visualization of connections between people.

What we learned

We learned a lot about data science, and learned to greatly appreciate how data extraction works. More specifically, we learned how to incorporate visualizations into Python to represent data as well as how to extract data from websites.

What's next for Finra People Search App

Hopefully we can incorporate the results of the python scripts (with full functionality) into a website or app to make the search engine more readily accessible.

Built With

Share this project:

Updates