Machine Learning in Microservices by Mohamed Abouahmed & Omar Ahmed

Author:Mohamed Abouahmed & Omar Ahmed
Language: eng
Format: epub
Publisher: Packt
Published: 2023-01-15T00:00:00+00:00

A common model we can use for unsupervised learning is an autoencoder. As mentioned in Chapter 4, an autoencoder is a neural network composed of an encoder and a decoder. The general purpose of an autoencoder is to take the data and compress it to a lower dimension similar to PCA. That way, it is able to learn the correlations and patterns between the different data features. Once it learns the patterns, it can feed the compressed data forward to the decoder where it tries to “recreate” the original data with what it has learned in the encoder stage.

While experts can study the data to determine what response times are considered an anomaly for a particular MSA, we can leverage machine learning to help us find patterns and relationships that may be hard to see even for an experienced developer.

With the learned parameters, we can then use this in our supervised regression models to achieve more accurate results when detecting anomalies and prevent false positives from occurring.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.