GTP | Devpost

Inspiration

We were inspired by our teaching assistant Jung Hyun who gave valuable insight on how optimization works.

What it does

Our project aims to optimize the pairing of moderators and advertisements to minimize negative impact and maximize positive impact on Tiktok. We have normalized variables in the advertisers data set to create a weighted composite priority score for the advertisers. In the moderator’s data set, we normalized variables to create a weighted composite capability score for the moderators. Using these two composite scores, we devised a more optimal matching between the advertisers and moderators.

How we built it

We build the optimizer using python, with jupyter notebook and libraries such as scipy, sklearn, and python matching.

Challenges we ran into

A challenge we encountered was the difficulty of deciding which variables we should consider in our composite score. Additionally, we faced some difficulty in matching the advertisements to the moderators as there were many factors to take into consideration.

Accomplishments that we're proud of

We have successfully calculated the composite scores using libraries that were new to us.

What we learned

We learned valuable skills while using python to work on the problem statement. Working on the hackathon as a group also allowed us to learn how to cooperate better as this was our first time working as a group.