Sytem to replicate Transaction Processing using BIG Data and Fraud Detection system using Machine Learning. ( WORK IN PROGRESS..... )
Big Data System to generate fake transaction data (credit & debit) and stream it to HIVE table using Apache Kafka.
Apache Spark will be used to process it during streaming.
Also working on building an anamoly detection model to flag and alert risky transactions that can be verified later for fraud.
logging events like health of microservices and training status of ML models
Monitor live system stats.
references: https://github.com/dbusteed/kafka-spark-streaming-example.git