Generating Synthetic Time-Series Data for Cow Activity Traces

Synthetic data allows studying and sharing potentially sensitive data without the risk of a privacy breach or leak. Synthetic data is generated by a model, which was trained on real data to learn its statistical properties. Thus, generating synthetic data can also be adapted and used for the prediction of missing values or detection of outliers. Training of such a generative model is usually done by using a Generative Adversarial Network (GAN) or a Variational Autoencoder (VAE). The goal of this thesis is to develop machine learning models for generating synthetic time-series sensor data for cow activity traces. Dairy cows are nowadays equipped with different sensors monitoring temperature, activity and other health related metrics. This data can be used to train a GAN or a VAE for generating similar sensor data. The generated data will then be used in several data analysis frameworks and the results compared to using real data directly

Download as PDF

Student Target Groups:

  • Students in ICE
  • Students in Computer Science

Thesis Type:

  • Master Thesis / Master Project

Goal and Tasks:

  • Thorough literature research on the topic
  • Implement a generative machine learning model
  • Evaluate the results of this generative model
  • Summarize the results in a written report, present and demonstrate the prototype

Recommended Prior Knowledge:

  • Creativity, interest in state-of-the-art machine learning methods
  • Programming skills in Python
  • Knowledge of in deep learning frameworks (TensorFlow/Keras or PyTorch) is recommended, but not mandatory


  • a.s.a.p.