Book Image

Actionable Insights with Amazon QuickSight

By : Manos Samatas
Book Image

Actionable Insights with Amazon QuickSight

By: Manos Samatas

Overview of this book

Amazon Quicksight is an exciting new visualization that rivals PowerBI and Tableau, bringing several exciting features to the table – but sadly, there aren’t many resources out there that can help you learn the ropes. This book seeks to remedy that with the help of an AWS-certified expert who will help you leverage its full capabilities. After learning QuickSight’s fundamental concepts and how to configure data sources, you’ll be introduced to the main analysis-building functionality of QuickSight to develop visuals and dashboards, and explore how to develop and share interactive dashboards with parameters and on-screen controls. You’ll dive into advanced filtering options with URL actions before learning how to set up alerts and scheduled reports. Next, you’ll familiarize yourself with the types of insights before getting to grips with adding ML insights such as forecasting capabilities, analyzing time series data, adding narratives, and outlier detection to your dashboards. You’ll also explore patterns to automate operations and look closer into the API actions that allow us to control settings. Finally, you’ll learn advanced topics such as embedded dashboards and multitenancy. By the end of this book, you’ll be well-versed with QuickSight’s BI and analytics functionalities that will help you create BI apps with ML capabilities.
Table of Contents (15 chapters)
Section 1: Introduction to Amazon QuickSight and the AWS Analytics Ecosystem
Section 2: Advanced Dashboarding and Insights
Section 3: Advanced Topics and Management

Using forecasting

Amazon QuickSight allows you to add forecasting to your dashboards without the need to develop complex ML models. To better understand how to configure forecasting, we will use the example dataset we configured in Chapter 2, Introduction to Amazon QuickSight.

Adding forecasting

For our example, let's assume that we need to develop a dashboard that contains forecasts about the total number of taxi fares in the future. As expected, our data has a certain degree of seasonality. Also, we can see from the line chart visual we developed in Chapter 3, Preparing Data with Amazon QuickSight, that during Sundays, there is a drop in the total taxi fares compared to the other days of the week. Identifying the most appropriate seasonality for our dataset is not always straightforward. In our example, we have different levels of seasonality depending on what time interval we will consider. A season can be 24 hours, or a week, or a year. Identifying the right seasonality...