Real-time and Predictive Insights on Vehicle Health and Driving Behavior Pattern Using Cortana Analytics

Play Real-time and Predictive Insights on Vehicle Health and Driving Behavior Pattern Using Cortana Analytics
Sign in to queue


Super computers have moved out of the lab and are now parked in our garage! These cutting-edge automobiles contain a myriad of sensors, giving them the ability to track and monitor millions of events every second. By 2020, 75% of these cars will be connected to the internet. Imagine having the ability to tap into this wealth of data generated by each car to provide best in class safety, reliability and personalized driving experiences for our families. Microsoft has made this imagination a reality via Cortana Analytics. Microsoft's Cortana Analytics is a fully managed big data and advanced analytics suite that enables you to transform your data into intelligent action.

Today we want to introduce you to Vehicle Telemetry Analysis Demo that we have built.

It leverages Event Hubs for ingesting millions of vehicle telemetry events into Azure and then uses Stream Analytics for gaining real-time insights on vehicle usage and persists that data into long-term storage for richer batch analytics. It takes advantage of Machine Learning for anomaly detection in real-time and batch processing to gain predictive insights. HDInsight is leveraged to transform data at scale, and Data Factory handles orchestration, scheduling, resource management and monitoring of the batch processing pipeline. Finally, Power BI gives this solution a rich dashboard for real-time data and predictive analytics visualizations.

In this demo, we will show you how to setup and run a sample Cortana analytics application to light up an Internet of Things (IoT) usecase related to vehicle telemetry analysis.



Download this episode

The Discussion

  • User profile image

    Nice presentation. Question: Is it possible to use Spark in this suite (or maybe new Prajna MS project) to do analysys? It works in memory so it's faster than Hadoop (used by hdinsight). Thanks.

Add Your 2 Cents