Serviços Personalizados
Journal
Artigo
Indicadores
- Citado por SciELO
- Acessos
Links relacionados
- Similares em SciELO
Compartilhar
Computación y Sistemas
versão On-line ISSN 2007-9737versão impressa ISSN 1405-5546
Resumo
PEREZ RAVE, Jorge; JARAMILLO ALVAREZ, Gloria e GONZALEZ ECHAVARRIA, Favián. Automated-Multitasking Methodology to Study Business Indicators using Data Science. Comp. y Sist. [online]. 2020, vol.24, n.3, pp.933-956. Epub 09-Jun-2021. ISSN 2007-9737. https://doi.org/10.13053/cys-24-3-3040.
The objective is to provide a multitasking methodology to study business indicators automatically using Data Science. This consists of 7 stages: data preparation, univariate analysis, bivariate analysis, grouping patterns, reduction of dimensions, supervised model training and validation. R, R-Studio (processing) and Rmarkdown (visualization) are used. The methodology is applied on four study cases (two from manufacturing sector and two from services) and provide useful information for firms and teaching-learning processes. The methodology distinguishes between media and response indicators, comprises more than 10 analysis methods, includes the three statistical scopes (univariate, bivariate and multivariate) and automatically answers six questions of interest for analysts.
Palavras-chave : Data science; business indicators; R-programming; analytics; data analysis; analysis methodology.