Anomaly Detection API by Azure Machine Learning

Anomaly Detection API is an example built with Microsoft Azure Machine Learning that can detect anomalies in time series data with numerical values that are uniformly spaced in time. The Anomaly Detection API is useful in several scenarios such as service monitoring by tracking KPIs over time, monitoring usage metrics such as #searches, #clicks, monitoring performance counters like memory, CPU, file reads, etc. over time.

Demo website

You can see the API in action on your dataset using the sample web application. It runs the Anomaly Detection API on your time series dataset and lets you visualize the data and any anomalies detected.


This project provides example C# source code for programmatically accessing the Anomaly Detection API from Azure Marketplace. You will need to sign up for the API before you can use it. The API offers free and paid tiers to use based on the need. See offer page for more details.

See the ADMarketplaceAPI for example of how to programmatically call the API. You will need to plug in your account key into the code to call the API successfully. The account key can be found under My Account on Marketplace.

Last edited Feb 11, 2016 at 11:20 PM by alokkirpal, version 8