Rice Datathon 2025 - Neurotech@Rice Track

• 

  Two graphs: one denoting the top 50 features in determining our Random Forest Model's decisions, and the other being a confusion matrix.

This is our submission for the Neurotech@Rice track for Rice Datathon 2025. Here, we analyze an EEG (electroencephalogram) dataset from people with psychiatric disorders (Park et al., 2021) using the Random Forest classification model to predict diagnoses of patients who have had an EEG.

At the end, we got 32.75% accuracy, which is good considering the time and data sample we were given.

Audio quality is somewhat spotty in the video, so please view the video description on YouTube for a complete transcript!

Built With

• eeg
• imblearn
• matplotlib
• pandas
• python
• scikit-learn