Diseño e implementación de una arquitectura en la nube para mitigar fraudes digitales en transacciones con tarjeta de crédito utilizando técnicas de Fast Data

Abstract

Today, driven by key business objectives in the new digital era, including competitive pressures, the need to capitalize on new market opportunities, the requirement for rapid data processing, and the generation of new, more complex, secure, and relevant customer experiences, there is a growing need for real-time data processing methods that protect the interests of organizations and the assets of their customers. Data streaming, also known as Fast Data, is not merely about retrieving data more quickly. It represents a fundamental shift in how we design data-centric applications. Many companies with Big Data systems have begun to move away from batch processing to keep pace with the third dimension of data: speed. While this is important, migrating to fast data architectures that enable scalable, fault-tolerant, and real-time systems is challenging. Therefore, this project proposes a data streaming method that benefits the Colombian banking sector (a fully cloud-based architecture), by designing and implementing the necessary steps to prevent fraudulent credit card transactions. This approach leverages real-time Big Data processing and algorithm development as its core strategy. The project aims to provide various organizations with information about transactional activity triggered by users of financial services, enabling them to take proactive measures against suspected fraud that could compromise customer assets and the reputation of financial institutions.