Automatic detection of relationships between banking operations using machine learning

In their daily business, bank branches should register their operations with several systems in order to share information with other branches and to have a central repository of records. In this way, information can be analysed and processed according to different requisites: fraud detection, accounting or legal requirements. Within this context, there is increasing use of big data and artificial intelligence techniques to improve customer experience. Our research focuses on detecting matches between bank operation records by means of applied intelligence techniques in a big data environment and business intelligence analytics. The business analytics function allows relationships to be established and comparisons to be made between variables from the bank's daily business. Finally, the results obtained show that the framework is able to detect relationships between banking operation records, starting from not homogeneous information and taking into account the large volume of data involved in the process. (C) 2019 Elsevier Inc. All rights reserved.