Salary Diviner

Landing Page
Home Page
Occupation Page
Final Page

Inspiration

We thought it would be really interesting to know our salary in advance. Also, we've never had experience with machine learning; therefore, we decided to make an iOS app that does it, powered by Face++ and Azure Machine Learning. We believe that this app can help people set expectation for their salary, no matter they are looking for entry-level positions or jumping from one industry to another.

What it does

With a photo taken and an occupation chose, this app would try to predict your salary based on your face.

How we built it

We used Face++ to make face recognition/detection, and Azure Machine Learning to predict salary by using Deep Forest Regression algorithm, mining Big Data salary statistics from U.S. Bureau of Labor Statistics.

Challenges we ran into

We are new to machine learning, and the entire thing is a challenge, from processing the data, to using various algorithms to judge accuracy of prediction models, to getting everything ready and deploy a Web API. For iOS, we have strong UI/UX tastes, so time was a big issue for us. We spent a lot of time at the beginning to do various prototyping (visual + interaction design), and then actually implement it.

Accomplishments that we're proud of

We learned everything within 36 hours, and make it work. We are proud of the entire thing. It just works!

What we learned

We learned various machine learning algorithms, as well as to learn to think like a data scientist.

What's next for Salary Diviner

We might upload it to App Store later after we catch time to add more features to it! For example, integration with Facebook would allow people to compare their "predicted "salaries with others. We think it would add a lot from the entertainment aspect.