Extração, transformação & carregamento melhorado: estudo de caso em aplicações financeiras

Abstract

Extract, Transformation & Load (ETL) is a three-step data integration model widely spread by the software community. This architecture aims to ensure a better visualization for data analysis. In this model data is extracted from one or more sources, transformed in a form that can be analyzed, and finally stored or loaded into a data warehouse, in a folder, or in another system. Due to the large growth in recent years in the number of individual investors on Brazil’s official stock exchange (B3), equity investments have become a topic much discussed by Brazilians and around the world. Operations like buying and selling assets in the short-term financial market, also know as trades, have become very popular among traders on the stock exchange. This work proposes the implementation of an algorithm to help short-term investors to make better decisions about purchases and sales of assets in the capital market. The application contains quantitative and qualitative analyses, which include the implementation of pie charts representing the users’ feelings towards the companies and/or their action code, in addition to the creation of graphs relating citations in Twitter by users and wheter they follow on this social media. Thus, the tool provides a visualization that makes it easier to analyse a particular company under study. Futhermore, this work is structed as follows: it presents an introduction to the ETC model; then, the concepts related to this model and the technologies applied to the algorithm creation are explored; finally, the methodology and development, the idea of the algorithm, and the results obtained from case studies this work validation are presented.