Um guia prático para apoiar tarefas preditivas em Ciência de Dados

Abstract

Currently, professionals from the most diverse areas need to explore their data repositories in order to extract knowledge and create new products or services. Several tools have been proposed in order to facilitate the tasks involved in the Data Science lifecycle. However, such tools require their users to have specific (and deep) knowledge in different areas of Computing and Statistics, making their use practically unfeasible for non-specialist professionals in Data Science. In this paper, we propose a guideline to support predictive tasks, more specifically, regression and classification. In addition, we present a tool, called DSAdvisor, which following the stages of the proposed guideline and aims to encourage non-expert users to build machine learning models to solve predictive tasks, extracting knowledge from their own data repositories. To evaluate DSAdvisor, we applied the System Usability Scale (SUS) questionnaire to measure aspects of usability in accordance with the user’s subjective assessment and the Net Promoter Score (NPS) method to measure user satisfaction and willingness to recommend it to others. This study involved 20 respondents who were divided into two groups, namely experts and non-expert users. The SUS method had a score of 68.5 which means a “good” product, and the results of using NPS get a value of 55% which means “very good” NPS.