The following are demo applications built using Hazelcast Jet. Each demo app is self-contained and showcase the data processing capabilities of Hazelcast Jet.
An application to analyze the real-time telemetry of commercial aircraft currently airbone using data from ADS-B transpoders. It computes noise levels and estimated CO2 emissions around major airports and urban areas as well as detecting aircraft taking off and landing through using a simple linear regression model. The results are then written to Graphite and rendered in a dashboard in Grafana.
An application to analyze tweet sentiments in real-time to compute cryptocurrency popularity trends. Tweets are streamed from Twitter and categorized by coin type(BTC, ETC, XRP, etc) and natural-language processing(NLP) sentiment analysis is applied to each Tweet to calculate the sentiment score. Jet aggregates scores from the last 30 seconds, last minute and last 5 minutes and prints resulting ranking table.
Extracts real-time vehicle data from the train simulation game (Open Transport Tycoon Deluxe) and analyses it using Hazelcast Jet. The analytical job predicts train collisions in real-time based on telemetry data supplied from the game. The prediction is pushed back to the running OpenTTD game to stop the affected trains.
TensorFlow is a popular library to train and use machine learning models. We integrate it with Jet to classify stream of events with the result of a TF model execution. This example uses the Large Movie Reviews Dataset as provided by the TensorFlow Keras Datasets and builds a model to predict whether a movie review is positive or negative.
Uses the webcam video stream of a laptop computer as a source and recognizes the objects using image recognition. The image classification is performed using a convolutional neural network pre-trained using a CIFAR-10 dataset.
A Markov Chain generator with probabilities based on supplied classical books. Markov Chain is a stochastic model describing a sequence of possible events in which the probability of each event depends only on the state attained in the previous event.
Shows how to use Jet for online machine learning use-cases. It combines real-time model training and prediction into one Jet pipeline to predict traffic patterns.
This example shows how Jet is used to spot the dramatically-named Death Cross for the price of Bitcoin, which is an indication to sell, Sell, SELL! The idea here is that we could automatically analyze stock market prices and use this information to guide our buying and selling decisions.
This application shows how to integrate a machine learning model generated with H2O with a Jet pipeline for real-time stream processing. We create an H2O Deep Learning model, train it with a sample data set (Breast Cancer Wisconsin Diagnostic Data Set) to prove statistical classification, export the model to a MOJO and incorporate it into a Jet Pipeline.
This example includes a Jet job that uses the CDC module of Jet to listen for changes on the configured inventory database and logs the events as they arrive.
This demo includes an example for Change Data Capture with Debezium, Kafka, MySQL and a Jet cluster inside Docker environment with Docker Compose. Debezium and Kafka Connect is used to and then published into a Kafka topic. The Hazelcast Jet pipeline listen for changes on the Kafka topic, logs the events as they arrive to the standard out and puts them to an IMap.
Stream changes using Change Data Capture, enrich the data, correlate (join) the records with other records and finally store the data into an Elasticsearch index, so an application can provide better functionality to the user.