Introduction to Anomaly Detector

Transcript

1. Microsoft.com/Learn
2. Introduction to Anomaly Detector Learn: https://aka.ms/gol-anomaly
3. Prerequisites  Basic familiarity with Azure  Ability to navigate the Azure portal  Basic familiarity with Azure Cognitive Services
4. Learning objectives  Describe the Anomaly Detector service  Briefly describe how Anomaly Detector works  Describe when to use Anomaly Detector
5. Agenda  Introduction  What is Anomaly Detector?  How Anomaly Detector works  When to use Anomaly Detector  Exercise
6. Agenda continued  Knowledge check  Summary
7. Introduction
8. Joke What do you call a Dragon that failed medical school?
9. Joke What do you call a Dragon that failed medical school? argon
10. Introduction Anomaly detection is an artificial intelligence technique used to determine whether values in a series are within expected parameters. Potential use cases include: • monitoring blood pressure • evaluating mean time between failures for hardware products • comparing month-over-month expenses for product costs
11. What is Anomaly Detector?
12. What is Anomaly Detector? Anomalies are values that are outside the expected values or range of values.
13. How Azure Anomaly Detector works
14. How Anomaly Detector works • The Anomaly Detector service identifies anomalies that exist outside the scope of a boundary. • The Anomaly Detector service accepts data in JSON format.
15. Data consistency recommendations If your data may have missing values in the sequence, consider the following recommendations. • Sampling occurs every few minutes and has less than 10% of the expected number of points missing. In this case, the impact should be negligible on the detection results. • If you have more than 10% missing, there are options to help "fill" the data set. Consider using a linear interpolation method to fill in the missing values and complete the data set. This will fill gaps with evenly distributed values.
16. When to use Anomaly Detector
17. When to use Anomaly Detector The Anomaly Detector service supports batch processing of time series data and last-point anomaly detection for real-time data.
18. Batch detection Batch detection involves applying the algorithm to an entire data series at one time. • Flat trend time series data with occasional spikes or dips • Seasonal time series data with occasional anomalies • Seasonality is considered to be a pattern in your data, that occurs at regular intervals. Examples would be hourly, daily, or monthly patterns. Using seasonal data, and specifying a period for that pattern, can help to reduce the latency in detection. • the maximum allowable temperature • the minimum allowable temperature • the acceptable duration of time for temperatures to be outside the safe range
19. Real-time detection Real-time detection uses streaming data by comparing previously seen data points to the last data point to determine if your latest one is an anomaly.
20. Exercise Anomaly Detector in Action
21. Exercise To test the capabilities of the Anomaly Detection service, we'll use a simple command-line application that runs in the Cloud Shell. • Create an Anomaly Detector resource • Run Cloud Shell • Configure and run a client application • Learn more
22. Knowledge check
23. Question 1 What is meant by seasonal data? A. Data based on the time or year it was recorded. B. How far apart the values are by default for each recorded period. C. Data occurring at regular intervals.
24. Question 1 What is meant by seasonal data? A. Data based on the time or year it was recorded. B. How far apart the values are by default for each recorded period. C. Data occurring at regular intervals.
25. Question 2 What is the purpose of specifying granularity in your JSON data object? A. It is used to indicate the recording pattern of the data. B. It tells the service how to chunk up the results that are returned for review, independent of the time series data pattern. C. It is used to indicate the range of acceptable values.
26. Question 2 What is the purpose of specifying granularity in your JSON data object? A. It is used to indicate the recording pattern of the data. B. It tells the service how to chunk up the results that are returned for review, independent of the time series data pattern. C. It is used to indicate the range of acceptable values.
27. Question 3 How does the Anomaly Detector service evaluate real-time data for anomalies? A. It collects all the values in a window of time and evaluates them all at once. B. It evaluates the current value against the previous value. C. It uses interpolation based on the current value and the previous value to predict what the expected value should be.
28. Question 3 How does the Anomaly Detector service evaluate real-time data for anomalies? A. It collects all the values in a window of time and evaluates them all at once. B. It evaluates the current value against the previous value. C. It uses interpolation based on the current value and the previous value to predict what the expected value should be.
29. Summary
30. Summary The Anomaly Detector detects anomalies automatically in time series data.
31. © Copyright Microsoft Corporation. All rights reserved.