Usage:

• First train the anomaly detection model, run the file:

model/train.py

• run kafka by docker

docker-compose up -d 

• ssh to the docker container

docker exec -it broker sh

• Create the required topics

kafka-topics --bootstrap-server broker:9092 --topic transactions --create --partitions 3 --replication-factor 1
kafka-topics --bootstrap-server broker:9092 --topic anomalies --create --partitions 3 --replication-factor 1

• Start the producer, run the file

python3 streaming/producer.py

• Start the anomalies detector, run the file

python3 streaming/anomalies_detector.py

• Send alerts to file alerts.txt,

python3 streaming/bot_alerts.py

• Tail the results

tail -f 100 alerts.txt