Inspiration
It was the only data science challenge.
What it does
The code tries to analyze and detect anomalies (outliers) from provided forensic data.
How we built it
We used Python for everything. At first, we made a brief analysis to get familiar with the dataset. Then, we implemented preprocessing and added DBSCAN algorithm.
Challenges we ran into
Interpretation of the dataset. Insufficient RAM. Evaluation of unsupervised learning.
Accomplishments that we're proud of
Solving the problem with insufficient RAM while using DBSCAN, which we solved by splitting the data into chunks. It kind of resembles cross validation.
What we learned
A machine.
Log in or sign up for Devpost to join the conversation.