Real-time data engineering for 360° customer intelligence in transaction banking

Live customer intelligence is becoming more and more important in transaction banking in the digital age, as new insights need to be derived from transaction data in real-time across banking channels. This paper presents an extensive study on the technologies, architectures, and performance aspects of real-time data engineering for 360° customer intelligence. They are able to compare the following technologies on ingestion, processing, orchestration, storage, and delivery of (？”/Web). exports): Apache Kafka, Flink, Airflow, Snowflake, GraphQL. Experiments demonstrate that these architectures are both useful and competitive in tasks like fraud detection, personalization, or churn prediction. According to the review, a certain insight into contemporary problems and anticipated research directions (such as federated data governance, serverless streaming, and autonomous pipeline optimization) is available. Financial engineers and researchers also need to design intelligent, compliant, and scalable banking systems by referring to the paper.