Rule-learning algorithms for biomarker identification

OData support
Supervisor:
Dr. Szűcs Gábor
Department of Telecommunications and Media Informatics

In biology, it is a great goal for scientists to be able to monitor cells and understand the processes inside. This way, significant consequences can be taken into consideration.

Nowadays, there are several types of techniques and procedures that allow experts to get closer to the solution e.g. taking images of the inside of live cells or representing cells as a network system. However, these models consist of numerous data points which result in time-consuming simulations and more difficult experiments. The purpose is to filter out the useful patterns and reject misleading ones to get a clear view.

In my thesis, I conduct literature search about the state-of-the-art in rule learning algorithms, and biomarker identification. I specify more general and widely used machine learning algorithms and complex techniques as well. The description of implementation and the results of my work are described and analysed in great details. During my explanation I highlight the arising problems and how I overcame them. Finally, I am convinced of the effectiveness of the results compared to the current working system.

Downloads

Please sign in to download the files of this thesis.