"Optimization and modeling of dynamic systems using pattern recognition and metaheuristic methods"

Machine Learning (ML) is a scientific discipline  of Computational Intelligence that is used in a wide range of applications such as pattern recognition and decision making. In recent years, machine learning techniques have been implemented to optimize systems on many levels. Machine learning techniques are applied in numerous fields in order to simulate and to solve problems such as energy consumption and big data analysis. The purpose of the PhD thesis is to study, develop and implement machine learning algorithms in modeling and prediction of time series.

Analysis of the subject of the Ph.D. thesis, provides the following research issues:

• Can we apply standard efficient and robust ML algorithms in time series forecasting? Can we apply them is dedicated problems?
• Can we propose and apply new ML algorithms in time series forecasting, having in mind dedicated problems?
• Can we propose and apply efficient meta-heuristic procedures in order to further enhance the time series forecasting problems?