-
Project dashboard screenshot
When determining what sorts of analysis we want to do, we looked around and saw an ad for a quick cash app. From there, we were inspired to look more into other ways of getting quick cash, such as predatory payday laons.
From there we chose to do the Business track, looking at the consumer complaints dataset and narrowing it down into that category of complaints. Attatching another dataset, the national well being survey from a few years ago, we were able to find out that taking out a payday loan severely impacts many aspects of your life.
From there, we were determined to create a tool that impacts people who are in a financial position where their next best option is a payday loan.
Some features of our tool that we are super proud of
- Text based Baysian model to read complaints and extrapolate where a company typically gets complaints
- XGBoost model that determines likelihood of snap eligbility
- Created loan estimator tool
Some major challenges we ran into
- Half of our group members have never used python for data science before
- Combining many tools into one that makes sense
- Collaborating when one persons work affects the other
- New to text based models
- Attempted to integrate third party site, without success (not all challenges lead to success)
Future direction
- We would like to build a third party integration that shows the user the ratings given by a
TEAM MEMBERS: Amishi Mishra, Erae Ko, Nancy Smart, Mihir Kale
Log in or sign up for Devpost to join the conversation.