Data science

Introduction

In a world where we are more and more connected, data is everywhere. You can extract knowledge and provide solutions benefiting society in a large variety of domains, including health, engineering, safety and security, business, and science. Methodologies for Data Engineering and Machine Learning are subjects of research and are constantly required to be improved to face new challenges brought by the increase of data complexity and new application scenarios. Most all off these fall under the broad umbrella of Artificial Intelligence (AI).

You can design or extend methods to mine, analyze, optimize, classify, or regress from data to solve a fundamental or applied research question of your choice. This method may follow a machine learning paradigm (e.g. supervised, unsupervised or reinforcement learning) or can belong to a related field, e.g. evolutionary computing. Problems in natural language processing, social networks, biometrics and computer vision, eHealth and medical related applications can be addressed. Concepts such as fairness, data quality, trust, and safety may also be considered.

Suggested Topics

Here is a list of current research topics, for inspiration:

Data engineering

  • Autonomous and robust extraction of information from the web and natural language text

  • Data integration and data cleaning

  • Online search engines and recommender algorithms

  • Process mining

  • Social media, social networks, and network science

Machine learning

  • Bayesian networks and probabilistic graphical models

  • Deep learning

  • Explainable AI

  • Evolutionary algorithms

  • Generative models

  • One-shot, transfer and multi-task learning

  • Reinforcement learning and multi-agent systems

  • Scalable neural networks

Biometrics and computer vision

  • Detection of manipulation in video and audio

  • Efficient searching of surveillance video

  • Finger vein, forensic face and biometric recognition

  • Mobile authentication

  • Systems fingerprinting for recognition and profiling

Further reading

A list of example project descriptions for this track: https://www.utwente.nl/en/eemcs/dmb/assignments/open/bachelor/

Information

For specific information on the content of this track, you may contact the track chairs: Decebal Mocanu and Nicola Strisciuglio.