1) “Combinatorial algorithms for variable selection in regression”, Erricos Kontoghiorghes, Professor, Cyprus Univ. of Technology, 2) “Statistical tests in topic modelling””, Ana Colubi, Professor, Univ. Oviedo, Spain – Τμήμα Μηχανικών Η/Υ και Πληροφορικής
Σας ενημερώνουμε για τις παρακάτω ομιλίες (double bill!) που θα δοθούν στα πλαίσια της σειράς εκδηλώσεων “Σεμινάριο CEID & Social Hour” και των ΔΠΜΣ ΥΔΑ, ΣΜΗΝ και ΟΣΥΛ.
CEID Seminar & Social Hour: 1) “Combinatorial algorithms for variable selection in regression”, Erricos Kontoghiorghes, Professor, Cyprus University of Technology and Birkbeck University of London, 2) “Statistical tests in topic modelling”, Ana Colubi, Professor, University of Oviedo, Spain.
Τίτλος: 1) Combinatorial algorithms for variable selection in regression, 2) Statistical tests in topic modelling.
Ομιλητές: 1) Erricos Kontoghiorghes, Professor, Cyprus University of Technology and Birkbeck University of London, 2) Ana Colubi, Professor, University of Oviedo, Spain.
Ημερομηνία-χώρος: Παρασκευή 1 Νοεμβρίου, 3-5μμ, ΤΜΗΥΠ, αίθουσα (θα ανακοινωθεί)
Περιλήψεις:
- (Joint work with Cristian Gatu, Marc Hofmann & Marios Demosthenous) Computational strategies for computing the best-subset regression models are The algorithms are based on a regression tree structure that generates all possible subset models. An efficient branch-and-bound algorithm that finds the best submodels without generating the entire tree is described. Approximate algorithms that improve the computational performance are investigated. These strategies are adapted to solve the problem of regression subset selection under the condition of non-negative coefficients. The solution is based on an alternative approach to quadratic programming that derives the non-negative least squares by solving the normal equations for a number of unrestricted least squares subproblems. The R package “lmSubsets” for regression subset selection is introduced and described. The package aims to provide a versatile tool for subset regression. Finally, the case of high-dimensional data where the number of variables exceeds the number of observations is considered. Within this context, a novel combinatorial solution is proposed. Experimental results are presented and analyzed.
- (joint work with Louisa Kontoghiorghes) A metric to quantify the relevance of specific subjects within a text is considered. The metric can be used to measure the statistical impact of a given text in related literature or to statistically compare topics between several corpora. To this aim, text mining tools are combined with Bayesian and frequentist statistical methods. First, topic modeling is suggested to identify relevant topics. The derived models are used to quantify the relative importance of a subject defined through a given set of terms or keywords by employing Bayesian techniques. Then, bootstrap tests are proposed to compare the prevalence of topics or subjects across different corpora. Illustrative empirical results are
Σχετικά με τους ομιλητές:
Ana Colubi is currently a visiting professor at the Justus Liebig University Giessen. She is a full professor at the University of Oviedo, Spain, on leave and visiting professor at King’s College London, UK, and Frederick University, Cyprus. She has published extensively on Probability Theory, methodological statistics, data analysis, ICT, , econometrics and environmental applications. She has co-organized several international conferences and was principal researcher in many research projects. She is the chair of the CRoNoS COST Action CA21163. She is a co-editor of Computational Statistics and Data Analysis and Econometrics and Statistics since 2015. She coordinates the CMStatistics and CFEnetwork (about 2000 and 1100 members respectively) and has been chairman of the European Board of Directors of the International Association for Statistical Computing, ERS-IASC (2014-2016).
Erricos Kontoghiorghes has obtained a Β.Sc., in Computer Science and Mathematics (1989) and a Ph.D. in Computer Science (1993) from Queen Mary College, University of London, UK. His research interests are within the interface area of computing and statistics. He is a professor at the Cyprus University of Technology since 2011. Previously he held faculty positions at the City University Business School, the Department of Computer Science, University of Neuchâtel, and the Department of Public and Business Administration, University of Cyprus. Since 2003 he also holds a visiting professorship at the School of Computing and Mathematical Sciences, Birkbeck, University of London, and since 2019 he is a visiting researcher at King’s College London. He is Co-Editor-in-Chief of the Journal Computational Statistics & Data Analysis (Elsevier) since 2000, Editor in Chief of the Journal Econometrics and Statistics (Elsevier) since 2015, and editorial board member of the Japanese Journal of Statistics and Data Science (Springer) since 2017.
