Associate Professor Mahdi Roozbeh
Faculty of Mathematics, Statistics and Computer Sciences, Semnan University, Iran
Title: Modeling extensive big data sets using robust penalized least-squares and support vector regression
Abstract:
Machine
learning is a state-of-the-art class of potent technologies in the modeling of
massive and big data sets that may address a wide range of important
problems that they may face in the modern world. The most popular of these
techniques for categorization is the support vector machine (SVM). Support
vector regression (SVR) is a technique for developing a regression model that
is an excellent addition to the machine learning family. These days, the
hardest problem to handle is wide big data sets. When working with wide big
data, the most difficult tasks are estimating the coefficients and interpreting
the results. High-dimension problems have many predictor variables, making it
impossible to apply traditional methods like the ordinary least-squares approach,
which works best when the main assumptions are met. One of the greatest
techniques for modeling wide big data sets is SVR. Having a good fit with great
accuracy is a very dependable and strong method. In this study, SVR is used on
a real wide big dataset about the gene expression in the production data of
riboflavin, and it is then compared with the well-known techniques of sparse
least trimmed squares and least absolute shrinkage and selection operator
(LASSO).
Keywords and
Phrases: Big data set, Multicollinearity, Outlier data, Robust
least trimmed squares.
Mathematics Subject
Classification (2020): 62J05,
90C11, 90C59
.
Biography:
Dr. Mahdi Roozbeh has been an associate professor in the Department of Mathematics, Statistics, and Computer Science department since 2018 at Semnan University. He graduated with a Ph.D. in Statistics from the Ferdowsi University of Mashhad in 2011, and he is an expert in regression modeling and analysis of high-dimensional data. He is the winner of the ISI (International Statistical Institute) Jan Tinbergen Award (the International Statistical Study Fund Foundation for young statisticians, and the prize for excellence in statistics is awarded every two years) in 2011 (which is held every 2 years in the world) for the best paper award during the World Statistical Congress in Dublin; the awards for Research Excellence at Semnan University in 2015, 2017, 2018, 2020, 2021, and 2023; the Prof. Behboodian Award; and was selected as the second young Iranian researcher in statistics by the Iranian Statistical Institute (2018). He was elected for the World Bank Trust Fund in the 62nd ISI WSC in Kuala Lumpur, 2019. His research focuses on high-dimensional deep learning, robust estimation, and semiparametric regression modeling, and he is an expert in machine learning and has been invited as the keynote speaker at some international world congresses and published more than 50 papers indexed by the ISI database and book chapters.
Reference
- https://scholar.google.com/citations?hl=en&user=KW9NMzoAAAAJ&view_op=list_works&sortby=pubdate
- https://www.scopus.com/authid/detail.uri?authorId=35590798700
- https://mahdiroozbeh.profile.semnan.ac.ir/#about_me
- https://www.researchgate.net/profile/Mahdi-Roozbeh
- https://www.webofscience.com/wos/author/record/1369569