Marie Skłodowska-Curie ITN - Researcher
Barcelona, Spain
thanasis.zoumpekas@ub.edu
Short bio
I was born in Xanthi, a town in the northern part of Greece, in 1993. I graduated from the 3rd Lyceum of Xanthi and later moved to Volos, Greece for my university studies. I received my Diploma and M.Sc. degree from the Department of Electrical and Computer Engineering, University of Thessaly, in 2018 and 2019 respectively. From May 2018 till September 2020, I was working as a Research Data Scientist and Machine Learning Engineer at the University of Thessaly. Currently I am a Marie Skłodowska-Curie ITN - Researcher at the Faculty of Mathematics and Computer Science, University of Barcelona. My research interests lie under the domain of data science. In particular, I am interested in machine learning, deep learning, artificial intelligence, big data analytics, visualizations, statistics, natural language processing and bioinformatics, not always in this order of significance.
Languages
Greek
Native Language
English
Proficient user - Proficiency (C2)
German
Independent user - Upper Intermediate (B2)
Machine Learning and interactive 3D visualisation of temporal point clouds for predicting morphological changes.
Providing Small and Medium-sized Enterprises (SMEs) with Business Intelligence through Machine Learning & Deep Learning.
Euclid Lab - Algorithms and Privacy Research Unit.
PhD Candidate / Faculty of Mathematics and Computer Science
Master of Science in Science and Technology of Electrical and Computer Engineering
Diploma of Engineering - Computer, Networks and Communications Engineering
Thanasis Zoumpekas, Elias Houstis, Manolis Vavalis. Science Direct
Thanasis Zoumpekas, Manolis Vavalis, Elias Houstis.
The purpose of this project was to compare innovativeness using the indicators from the European Innovation Scoreboard. We compared the scores of Greece with the EU average scores over the period 2010-2017. We analysed systematic overand underperformance of Greece and the trends of these indicators over the years. Furthermore we used machine learning techniques to determine the most important features that drive the fluctuation of summary innovation score on EU and Greece level.
The purpose of this project was to provide a comprehensive analysis of the cryptocurrency market. We used several methods of statistics and machine learning. We utilized deep learning algorithms for prediction. In particular, multiple neural network architectures are developed and evaluated. The final product was a real-time prediction software of closing price of Ethereum.
An individual semester project. Inequality joins across distributed databases are implemented using the Apache Spark, under the broader concept of Map-Reduce.
A team semester project. Fundamental techniques of speech and audio processing are utilized to extract and create basic signal features. Phoneme and speech classification has been conducted using probabilistic and statistical modeling.