https://towardsdatascience.com/building-a-realtime-nlp-pipeline-using-kafka-and-spacy-d4ad636be702 > In this tutorial we will build a realtime pipeline using Confluent Kafka, python and a pre-trained NLP library called spaCy. > As prerequisites we should have installed docker locally, as we will run the kafka cluster on our machine, and also the python packages spaCy and confluent_kafka -pip install spacy confluent_kafka.