Znanstveni centar izvrsnosti za znanost o podatcima i kooperativne sustave

Centar ACROSS-DataScience djeluje kao središte suradnje akademske zajednice s poslovnim i javnim sektorom u istraživanju i razvoju novih metodologija i naprednih inženjerskih pristupa za znanost o podatcima i kooperativne sustave.

Poziv na predavanje: „Statistical...

Istraživačka jedinica za znanost o podatcima Centra izvrsnosti za znanost o podatcima i napredne kooperativne sustave, u okviru projekta "DATACROSS – Napredne metode i tehnologije u znanosti o podatcima i kooperativnim sustavima", organizira predavanje istraživačkog seminara

Statistical analysis and complexity reduction of unevenly spaced data in orthogonal bases: Theory and applications

koje će održati prof. dr. sc. Sergej Prokhorov sa Samara National Research University, Rusija. Predavanje će se održati u petak 19. 07. 2019. u 11:00 sati na Institutu Ruđer Bošković u predavaonici I krila (krilo Ivan Supek).

Više o predavaču i predavanju možete pročitati u opširnijem sadržaju obavijesti.

Sažetak: Many methods and algorithms were proposed to deal with evenly spaced data. In practice, we often have to face unevenly spaced data due to the presence of missing values, sensors failures, the nature of data and so on. Most approaches to the unevenness issue suggest embedding incomplete data. However, this may result in larger errors if computed embeddings are not suitable. The present study is an overview of methods and algorithms for handling unevenly spaced data with the orthogonal bases. Building regression models with a family of orthogonal bases permits to propose a new interdisciplinary approach to statistical analysis and complexity reduction of unevenly spaced data by applying the theory of Fourier and orthogonal series. 

The lecture covers the following topics:
1.    A theoretical and empirical analysis of unevenly spaced random sequences and time series. 
2.    Modeling unevenly spaced time sequences with given properties.
3.    Reducing the complexity of unevenly spaced time series models with orthogonal bases.
4.    Applications of methods and algorithms of unevenly spaced time series analysis in natural and engineering sciences. 


Biografija: Sergej Prokhorov defended his PhD in Information Measurement Systems in 1974 and received his Dr.Sci. in Information Measurement Systems in 1987. He became a full professor in 1988. He worked as the Dean of the Faculty of Computer Science at Samara State Aerospace University in 1988-2005 and the Director of Samara Branch of Information Systems Institute in 1993-2000. He was the President of the Head Council of the Ministry of Higher Education of Russia for the automation of scientific research in 1988-1996. He has been holding the position of the Head of Information Systems and Technologies Department at Samara National Research University (formerly Samara State Aerospace University) since 1988 and has been working as the Member of the Higher Attestation Commission of Russian Federation since 2014.

He was a thesis mentor for 4 Dr.Sci. researchers, 27 PhD students, 3 Master’s students, 154 Higher Education Diploma’s students as well as a co-mentor for 5 PhD students. He is an author of more than 485 refereed scientific papers published in Russian and in English. Among them are 23 monographs and 61 patents. He took part in 132 International, All Russian, and Regional conferences. His research interests are in the areas of signal processing, statistical data processing, statistical learning, and automation control systems.

He is a full member of the International Informatization Academy, the International Academy of Navigation and Motion Control, the International Academy of Telecommunications and Informatics as well as a corresponding member of the Russian Academy of Natural Sciences. He is the Merited Worker of Higher Education of the Russian Federation. He also is awarded for the medals of  M Keldysh, Y Gagarin, S Korolev from the Federation of Cosmonautics of the Russian Federation.

Popis obavijesti