Event Details

Professor Vittorio Maniezzo, University of Bologna will present a seminar to the Management Science Department

Abstract: Time series analysis plays a critical role in data analytics, an effective modeling of nonlinear trends is essential for obtaining actionable results, notably for forecasting and missing values imputation. The segmentation of time series and the corresponding detection of change points stand out for their practical implications. This paper presents preliminary results of a study on the applicability of mathematical programming, and in particular matheuristics, to time series segmentation.

Presenter: Vittorio Maniezzo vittorio.maniezzo@unibo.it

Bio: Vittorio Maniezzo is full professor in computer science, at the Computer Science and Engineering Department of the University of Bologna. He has been working on heuristic algorithms for combinatorial optimization for decades, first contributing to the original design of Ant System optimization, then bringing forth the area of matheuristics. He is currently interested in operational analytics, combining matheuristics with predictive modules.

He graduated in Electronics and Computer Science Engineering from the Polytechnic of Milano, where he also obtained a Ph.D. in Computer Science Engineering. He then won a fellowship at the Joint Research Center of the European Commission (Ispra). His academic career went through the positions of researcher, associate professor then full professor, always at the Cesena campus of the University of Bologna and always with the affiliation to the Department of Computer Science of this same University. Among his research achievements, he was one of the original designers of the Ant System algorithm – later evolved as Ant Colony Optimization (ACO). More recently, he has been one of the promoters of the active community on matheuritics (2006 to present). Both contributions have given rise to a series of specific, currently active, international workshops. He is the author of more than a hundred scientific works, including over 40 listed in the scopus or ISI databases.