Activity Data Generation for Smart Home Applications

Motivation.

The current and future development of intelligent capabilities in smart home systems rely on machine learning methods, which in general require of large amounts of training data. However, the collection of real-world data is expensive, time-consuming, and in general hard to come by.

Goal.

Devise and implement a machine learning approach to automatically generate synthetic data based on real-world data.

Requirements.

- Eagerness to learn and work hard

- Programming experience

- Some hands-on experience in machine learning