Please use this identifier to cite or link to this item: http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/1521
Title: Recent Developments in Data Science: Comparing Linear, Ridge and Lasso Regressions Techniques Using Wine Data
Authors: Mayooran, T.
Gabirial, M.
Rahman, A.
Keywords: Standard regression;Ridge, Lasso
Issue Date: 2019
Publisher: University of Oxford UK
Citation: Thevaraja, M., Rahman, A., & Gabirial, M. (2019, April). Recent developments in data science: Comparing linear, ridge and lasso regressions techniques using wine data. In Proceedings of the international conference on digital image & signal processing (pp. 1-6). University of Oxford UK.
Abstract: Big data is the reality of the 21st century. However, big data modeling and prediction require advanced level analytics which encompasses both the computing-intensive and statistics-oriented analysis tools in data science. Regression analysis is the statistical method for predictive modeling, and it is one of the most commonly used methods in many scientific fields such as engineering, the physical and chemical sciences, economics, management, life and biological sciences, and the social sciences, sociology, geology, etc. Satisfying the assumptions such as collinearity between variables ought to be a significant issue in data science. Advanced level tools such as Lasso and Ridge regression methods are designed to overcome such problem. In this study we discussed about comparing linear regression with the Ridge and Lasso. The Vinho Verde white wine test data from the Minho (northwest) region of Portugal is used to analyze advantages of each of the three regression analysis methods. All the required calculations and graphical displays are performed using the R software for statistical computing.
URI: http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/1521
ISBN: 978-1-912532-09-4
Appears in Collections:Interdisciplinary Studies

Files in This Item:
File Description SizeFormat 
Recent_Developments_in_Data_Science.pdf16.66 kBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.