Mapping of Challenges and Methods for Data Science Application

Resumo

Context. Data Science is described as an interdisciplinary field combining statistics, mathematics, computer science, and economics to extract valuable knowledge and insights from data. The article discusses the differences between Data Science and conventional statistics, the importance of Big Data, and the techniques used to analyze large volumes of data. The emerging profession of Data Scientist is also addressed, highlighting the skills needed to work with complex, unstructured data. Objective. The objective of this article was to characterize the challenges and methodologies of Data Science applications. Methodology. The methodology of this study involved a systematic mapping of the literature, providing a comprehensive overview of Data Science and highlighting development methods with a focus on reproducibility, transparency, and strategic alignment. The differences between Data Science and conventional statistics were analyzed to better understand the evolution of the field and the competencies required of professionals. Results. Of the 29 articles retrieved from scientific databases, 4 met the inclusion and exclusion criteria. It was observed that 50% of the publications occurred in 2022, indicating a growing interest from researchers in the field. Conferences represented the primary publication format, accounting for 75% of the works, while journals accounted for 25%. Conclusion. The main challenges identified include the difficulty in ensuring the reproducibility of studies, the misalignment between projects and organizational goals, and the lack of standards and frameworks for the Data Science project lifecycle. The research suggests the need for investment in education and training, as well as more methodologies to evaluate the impact of projects on business, aiming for greater return on investment.