Data-nator

UX/UI, Mobile app design


Summary
A mobile solution to label data for machine learning models.

Problem
In order to train a machine learning model, a large amount of data needs to be labeled. For small scale data science teams or data scientists that wish to label their own data, this app makes it easy, convenient and collaborative.

The app using swipes to quickly assign labels (up to four). The user uploads their data, sets up a question for labeling and can even add contributors to help annotate their data. Future iterations include annotating images.


Left: Data-nator mobile app


A Collaborative Approach
With Data-nator, the data labeling responsibilities can be shared amongst data scientists, to work together on labeling an entire data set. This is an alternative to Amazon's Mechanical Turk or other data labeling service.