HabitHacker

What it does

HabitHacker uses a machine learning model to determine your mood based on your daily journal notes. It then compares this to the music you listen to and the weather. Then, HabitHacker is able to show you patterns in your mood based of your surroundings, the weather and music.

How we built it

We trained our own machine learning model with over 85,000 data points, and a 90% accuracy rate. We used a TFIDF-Vectorizer, Label Encoder and Logistic Regression Model in scikit-learn.

Challenges we ran into

Training times for the model. We had to retrain many times to ensure the best accuracy rate. Setting up APIs. Building a GUI.

Accomplishments that we're proud of

Overall, we're just proud we were able to build a whole app in 2 days, and on top of that, an app that can teach you something about yourself.

What we learned

We've learnt more about working with APIs, and how to organize projects for tight deadlines.

What's next for HabitHacker

We plan to add more data funnels for things that could alter your mood such as screen time. This way, the user could get even more insights.

Built With

• kivy
• numpy
• pandas
• python
• scikit-learn
• spotipy