ODT - THESIS TOPIC: Zsolt Csaba Johanyák: Fuzzy model generation from ...


	Login Forum Organisation officers member institutions resolutions (in Hung.) by-laws (in Hung.) regulations Doctoral program doctoral schools by institution by research field by number ceased working academic staff member by name by research topic archives thesis topic proposals by doctoral school by research field by research topic supervision of PhD students search students Theses defended by institution by research field by name by date by doctoral school Naturalized degrees by institution by research field by name by date Habilitation presentations by institution by branch of science by name Documents (all in Hung.) public documents news legal backgrounds guidelines Contacts contact details partners

Thesis topic proposal

Fuzzy model generation from large data samples

THESIS TOPIC PROPOSAL

Institute: University of Miskolc
computer sciences
JÓZSEF HATVANY DOCTORAL SCHOOL FOR COMPUTER SCIENCE AND ENGINEERING

Thesis supervisor: Zsolt Csaba Johanyák
Location of studies (in Hungarian): Informatikai Intézet
Abbreviation of location of studies: INF

Description of the research topic:

Automatic fuzzy model generation from sample data is relatively well supported by scientifical articles and commercial software products (e.g. Matlab’s Fuzzy Logic Toolbox) in case of low number of dimensions and small data samples. However, with the increase of the dimensionality of the problems (e.g. more than 5-7 input dimensions) and the increase of the amount of the available sample data (e.g. more than 200 samples) the determination of the optimal system parameters can become extremely hard and time consuming.

Through this project, we aim to develop new techniques that enable us to efficiently create and optimize fuzzy systems for cases when the number of input dimensions exceeds 25 and the available data can contain more than 10 000 samples. The research may include deep learning like techniques, nature inspired optimization techniques, data analysis (e.g. ANOVA, SVD), as well as fuzzy rule interpolation.

Deadline for application: 2021-02-01

2024. IV. 17.
ODT ülés
Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).