Kafka-ML: Connecting the data stream with ML/AI frameworks

Machine Learning (ML) and Artificial Intelligence (AI) depend on data sources to train, improve, make predictions through their algorithms. With the digital revolution current paradigms like Internet of Things, this information is turning from static continuous streams. However, most ML/AI frameworks used nowadays are not fully prepared for revolution. In paper, we propose Kafka-ML, a novel open-source framework that enables management pipelines Kafka-ML provides an accessible user-friendly Web user interface where users can easily define ML models, then evaluate, deploy them inferences. itself components it deploys managed containerization technologies, which ensure portability, easy distribution, other features such as fault-tolerance high availability. Finally, approach has been introduced manage reuse streams, may eliminate need storage or file systems.