TLDR:
We use sets of machine learning models to deliver actionable financial insights to your inbox every morning.
Inspiration
We know plenty of individuals that are scared of investing, and really are unsure where to start. After taking that first step and getting interested, following the news and understanding what news matters is hard. We provide a simple user interface to allow for anyone to receive daily news updates.
What it does
Users input companies they'd like to follow, and we deliver actionable news to their inbox.
How we built it
We built it using Google Cloud Platform AI+ML, Firebase Hosting, Firebase Functions, and Firebase Realtime Database. We _ extensively _ used machine learning - we apply online clustering to minimize noise and help remove duplicate news articles for individuals. In addition, clustering is used to ensure validity of articles by helping rank importance through cohesion. We leveraged natural language semantics to provide a one-liner summary on a per article basis, given the article and other similar articles. We used sentiment analysis to determine article outlook.
Challenges we ran into
Our machine learning models were hard to work with due to the absolute breadth of news data. We also had to build support for a wide range of companies, and the software to support that was not trivial.
Accomplishments that we're proud of
We're proud of our multi-tiered machine learning models to provide actionable financial insights.
Built With
- clustering
- firebase
- javascript
- machine-learning
- python
Log in or sign up for Devpost to join the conversation.