Book Image

Time Series Analysis on AWS

By : Michaël Hoarau
Book Image

Time Series Analysis on AWS

By: Michaël Hoarau

Overview of this book

Being a business analyst and data scientist, you have to use many algorithms and approaches to prepare, process, and build ML-based applications by leveraging time series data, but you face common problems, such as not knowing which algorithm to choose or how to combine and interpret them. Amazon Web Services (AWS) provides numerous services to help you build applications fueled by artificial intelligence (AI) capabilities. This book helps you get to grips with three AWS AI/ML-managed services to enable you to deliver your desired business outcomes. The book begins with Amazon Forecast, where you’ll discover how to use time series forecasting, leveraging sophisticated statistical and machine learning algorithms to deliver business outcomes accurately. You’ll then learn to use Amazon Lookout for Equipment to build multivariate time series anomaly detection models geared toward industrial equipment and understand how it provides valuable insights to reinforce teams focused on predictive maintenance and predictive quality use cases. In the last chapters, you’ll explore Amazon Lookout for Metrics, and automatically detect and diagnose outliers in your business and operational data. By the end of this AWS book, you’ll have understood how to use the three AWS AI services effectively to perform time series analysis.
Table of Contents (20 chapters)
1
Section 1: Analyzing Time Series and Delivering Highly Accurate Forecasts with Amazon Forecast
9
Section 2: Detecting Abnormal Behavior in Multivariate Time Series with Amazon Lookout for Equipment
15
Section 3: Detecting Anomalies in Business Metrics with Amazon Lookout for Metrics

Using a trained model

The following diagram illustrates how Amazon Lookout for Equipment works at inference time:

Figure 11.1 – Inference scheduler overview

Let's dive into the different steps of this pipeline:

  1. New time series are generated: depending on your use case, you might collect new sensor data directly from your piece of equipment or directly access a piece of software such as a historian.
  2. As you did at training time, you will need to push this fresh data to a location on Amazon S3 (you will see in the following Configuring a scheduler section how this location is configured).
  3. Your inference scheduler will be configured to run regularly (for instance, every five minutes or every hour). Each time it wakes up, it will look for fresh data and run it against your trained model. Once the model generates new results, the scheduler will store it in JSON format.
  4. At the end of each scheduler run, the inference results will be...