Book Image

Machine Learning with the Elastic Stack - Second Edition

By : Rich Collier, Camilla Montonen, Bahaaldine Azarmi
5 (1)
Book Image

Machine Learning with the Elastic Stack - Second Edition

5 (1)
By: Rich Collier, Camilla Montonen, Bahaaldine Azarmi

Overview of this book

Elastic Stack, previously known as the ELK stack, is a log analysis solution that helps users ingest, process, and analyze search data effectively. With the addition of machine learning, a key commercial feature, the Elastic Stack makes this process even more efficient. This updated second edition of Machine Learning with the Elastic Stack provides a comprehensive overview of Elastic Stack's machine learning features for both time series data analysis as well as for classification, regression, and outlier detection. The book starts by explaining machine learning concepts in an intuitive way. You'll then perform time series analysis on different types of data, such as log files, network flows, application metrics, and financial data. As you progress through the chapters, you'll deploy machine learning within Elastic Stack for logging, security, and metrics. Finally, you'll discover how data frame analysis opens up a whole new set of use cases that machine learning can help you with. By the end of this Elastic Stack book, you'll have hands-on machine learning and Elastic Stack experience, along with the knowledge you need to incorporate machine learning in your distributed search and data analysis platform.
Table of Contents (19 chapters)
1
Section 1 – Getting Started with Machine Learning with Elastic Stack
4
Section 2 – Time Series Analysis – Anomaly Detection and Forecasting
11
Section 3 – Data Frame Analysis

Using one-sided functions to your advantage

Many people realize the usefulness of one-sided functions in ML, such as low_count and high_mean, to allow for the detection of anomalies only on the high side or on the low side. This is useful when you only care about a drop in revenue or a spike in response time.

However, when you care about deviations in both directions, you are often inclined to use just the regular function (such as count or mean). However, on some datasets, it is more optimal to use both the high and low versions of the function as two separate detectors. Why is this the case and under what conditions, you might ask?

The condition where this makes sense is when the dynamic range of the possible deviations is asymmetrical. In other words, the magnitude of potential spikes in the data is far, far bigger than the magnitude of the potential drops, possibly because the count or sum of something cannot be less than zero. Let's look at the following screenshot...