ANALYSIS INFERENTIAL NO PARAMETRIC IN ENVIRONMENTS COMPUTING IN CLOUD: EFFECTIVENESS DO GOOGLE COLAB FOR RESEARCHERS

This research addresses the application of non-parametric inferential statistical analysis in large volumes of data, using Google Colab as a cloud computing platform. In the context of a growing interest in accessible analysis tools, the problem focuses on the need for environments that allow complex analyses to be performed without assuming data normality. The general objective was to investigate the effectiveness of Google Colab in performing nonparametric tests and its integration with machine learning techniques. The Giftedean neoperspectivist paradigm, using complexity, critical and constructivist theories, with the application of the hypothetical-deductive method to test the formulated hypotheses. The techniques included Mann-Whitney, Kruskal-Wallis and Wilcoxon analyses, as well as supervised learning, using Python. The main findings revealed the efficiency of Google Colab in analyzing asymmetric data, highlighting its ability to integrate statistical methods with predictive models. Gaps include comparison with other platforms and the exploration of more advanced AI techniques. The theoretical contributions reinforce the use of nonparametric methods in big data; methodologically, it offers a replicable framework; and empirically, it demonstrates the viability of accessible tools for data analysis. The added value lies in the democratization of access to analysis technologies and in the enhancement of studies in different areas of knowledge.