Datamining Tool: Multiple Regression and Logistic Regression in a Web Platform of a Datawarehouse

Abstract:

This article summarizes the works about the design of a datamining tool based on statistical methods (Multiple Regression and Logistic Regression). Dataminig [1] is a science which uses mathematical and statistical methods to highlight the links hidden between two or several different variables. This science helps the leaders in their forecasts and in their decision-making. Among the numerous statistical methods used by datamining, only the multiple regression and the logistic regression are objects of a complete study in this work. Indeed, the Multiple Regression is the first mathematical method used in the implementation. It allows to clarify the variation of a variable with regard to the other variables apparently independent. However, the multiple regression gives answers in terms of numerical values such as 0 or 1. The logistic regression allows to clarify the variation of a binary variable (0 or 1), Exist or not according to the other variables. So, the logistic regression is an extension of the multiple regression. After the study of these methods, an implementation on a web platform (approachable application via a web browser) is proposed. Indeed, this IT application is intended to process millions of data for a company and to make simulations).

• Periodical: Advanced Materials Research (Volumes 694-697)
• Pages: 2299-2307
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.694-697.2299
• Online since: May 2013
• Authors: Fatou Faye*, Mbaye Sene
• Keywords: Data Mining (DM), Datawarehouse, Logistic Regression, Multiple Regression, Web Platform

References

[1] G. CELEUX, E.DIDAY, Classification automatique des données, environnement statistique et informatique, Dunod (1989). Definition of the datamining on http://www.ultra-fluide.com/savoir-faire/datamining.htm

[2] Lebart L., Morineau A., Piron M., Statistique exploratoire multidimensionnelle. Dunod (1995) Saporta G., Probabilités, analyse des données et statistique. Dunod(1990) Definition of the ACP (Analyze in Main Components) on http://www.aiaccess.net/French/Glossaires/GlosMod/f_gm_analyse_composantes_principales.htm

DOI: 10.51257/a-v1-af620

[3] Definition of Decision trees on http://fr.wikipedia.org/wiki/Arbre_de_d%C3%A9cision

[4] Definition of the networks of neurons on http://fr.wikipedia.org/wiki/R%C3%A9seau_de_neurones_artificiels