Skip to content

Latest commit

 

History

History
53 lines (33 loc) · 2.63 KB

README.md

File metadata and controls

53 lines (33 loc) · 2.63 KB

A Self Financial Wellness Assessment Tool (like a financial selfie)

2nd place at Breaking Barriers Buildathon, by MIT FinTech Club, FINRA, and Deloitte

Project Team: Nidhi Sharma, Kiran ChandreGowda, Mallika Saidi, Diego Benitez Concha, Yasmin Omrani, and Brian Vannah

Advisors and Friends: Ada ZhouYing, Peter Smiley, Syed J. Zaidi, and Stephen Cohen

Project Information

Check out the Solution Summary and Submission PDF for our design process and theoretical approach. For the social media portion of our project, we predicted financial risk tolerance given text as input. After training a model using the (MBTI) Myers-Briggs Personality Type Dataset from Kaggle, we used people's public Twitter accounts (specifically, the text in their tweets) to predict their Myers-Briggs personality type. We then used a mapping we found on the internet that correlated personality types to levels of financial risk (see diagrams below).

project flowchart

project diagram

personality type table

Run it Yourself

This repo only contains the code for about half of our project (using Python NLP to convert from Twitter to financial risk). Once you run the code, it will ask you for a Twitter handle (which must be public or followed by the Twitter API user). For example, I would type in mcuban to get information about Mark Cuban. About 20 seconds later (depending on your computer and internet speed), there will be a pop up box with the predicted Myers-Briggs personality type and level of risk tolerance.

example input

example output

1. Download the code.

Clone the repository, download the zip file, whatever.

2. Get a Twitter API key

The file twitter_access.py needs four API keys. Find the part in the file where it declares consumer_key, consumer_secret, access_token, and access_token_secret, and put your keys/tokens in there. You can apply for a Twitter development account (or ask a friend to borrow theirs but don't commit fraud).

3. Install the dependencies

You need Python 3, and a bunch of Python modules. My recommendation for the modules is to just install pip, and then try to run the main file (get_risk.py) and use pip to install the modules that it asks for.

4. Run the main file

Execute get_risk.py however you execute python scripts, usually something like python3 get_risk.py in terminal, or run it using your IDE of choice (IDLE or Spyder for many people).



Shoot me a text, email, etc. if you have questions. -Brian