The inspiration was the usual student's life: Students have to balance between social life, alcohol, and of course trying to get high grades. It'd be nice to have a personal assistant or counselor that helps students to keep them in track for reaching their study goals.
What it does
Our app makes prediction about the grades you will get, based on the amount of sleep you got and the amount of alcohol you consumed. It applies a machine learning algorithm on the dataset of all students to make a personalised prediction.
How we built it
Frontend: We use Ionic 1 + d3js for the front-end app. In addition to that we used the typical web technologies: JS, HTML, CSS, JSON. Backend: For the machine learning functionality we used TensorFlow. Consequently we wrote the backend in Python. We use PostgreSQL as our database interface it through the SqlAlchemy ORM. The Backend is hosted on an Azure VM (running Ubuntu).
Challenges we ran into
New and unfamiliar technologies like Azure and Machine Learning (also relearning Python...) itself were the most challenging part for our team. Also we were confront with the team-building aspect and the responsibility of distributing the work.
Accomplishments that we're proud of
We grew as a team and we learned new technologies from scratch. The app looks nice and it works just as planned - that's cool.
What we learned
It's important to remain focused when implementing a product idea. We also got known some frameworks and new hype technologies like cloud computing and machine learning.
What's next for Dreamwalker
First of all we hope to remain friends even after hackaTum. Even though we have a lot of good ideas how to improve our product, we aren't sure whether we will continue the project.
- Python 3
- Ionic 1