Project X-Ray

The Problem

We are responding to the third prompt from the San Diego Zoo Hackathon 2018: Trafficking on Social Media. Social media is a hotbed for the sale of illegally poached animals and their associated products all over the world. Often, sellers reach out to the public in hopes of finding a buyer interested in their wares. Many social media companies take preventative measures against these illicit dealings, but the algorithms used have loopholes and do not preserve the tweets for deeper analysis. Our hope is that our platform will allow conservationists to combat these loopholes.

Our Proposal

Our solution is a platform for researchers and conservationists to go further than social media’s current efforts to fight animal trafficking and sale. We allow a user to search recent tweets with a variety of keywords and helper phrases. While looking through tweets, our algorithm gives each tweet a point score to determine how suspicious it is. If the tweet earns enough points, it will be returned to the expert for review. We also track the location of potentially malicious tweets on a heat map to help the conservationist understand where illegal trafficking activities are most popular. The results obtained can be given to law enforcement as well as used to set up fake buyer accounts. This allows conservationists to be better informed about the animal trafficking industry and its evolution and as well as provide critical information for pursuit of these criminals. Front end: https://github.com/ShrimpMantis/zooHackathon Back end: https://github.com/googolplex8/Zoo-Hackathon

We Assume that...

1. We assume the posts made regarding the sale of illicit goods are written in english and public

2. there are people to work the platform and interact with the tweets selected by our algorithm.

Constraints to Overcome

The biggest barrier to entry was the obtainment of keys needed to access the Twitter, Facebook, Instagram, and Reddit APIs. We were only able to get ahold of the Twitter key because we had a copy of it before the hackathon began from a prior project. This allows us to access public tweets in real time, as well as the mountains of metadata points associated with each tweet.

Current Work

Adapt the dredger to work on multiple social media sites, create a neural network so the algorithm can better weight scores, include image analysis

Current Needs