Search

Big Data Analytics


Description Contemporary deep learning topologies such as end-to-end learning and representation learning systems require substantially more data than conventional machine learning approaches. Therefore, research fields that relies heavily on data that is laborious and costly to collect and annotate, struggle to procure enough reliably labelled instances to adequately train such systems.
Task In this thesis, the student(s) will be extending CAS²T, a java-based complex networks analyser toolkit for extracting big data from social multimedia.
Utilises CAS²T
Requirements Preliminary knowledge in Machine Learning, Good programming skills (e.g. Java, Python).
Languages English and German.
Supervisor Shahin Amiriparian, M. Sc. (shahin.amiriparian@informatik.uni-augsburg.de)