Amazon Quicksight is a business intelligence service infused with machine learning functionalities, intended for cloud computing, which is a part of the Amazon Web Services solutions. As with any business intelligence tool, the ultimate goal of Quicksight is to extract useful insights from data to support the decision-making process. The final product of the Quicksight service for end users are reports and dashboards which are often enhanced with machine learning functionalities, and can easily be incorporated into existing applications and web portals.
Machine learning capabilities
Like many other business intelligence tools, Amazon Quicksight contains many machine learning (ML) functionalities. Quicksight uses the robust Amazon Web Services ML functionalities to enable deeper analysis of business data with machine learning and natural language processing. In the next part of the article we will briefly explain forecasting, anomaly detection, the creation of automatic narratives and creation of visualizations through natural language processing functionalities.
Forecasting – Quicksight provides users with the ability to make predictions regardless of their level of technical knowledge. With just a few clicks, the tool uses the Random Cut Forest algorithm to make predictions, taking into account complex problems such as seasonality and trends, outliers and missing values.
Anomaly detection – anomaly detection enables continuous data analysis by a large number of indicators for the purpose of finding anomalies, i.e. outliers and revealing hidden trends in the data.
Autonarratives – autonarratives provide a slightly different way of analyzing data compared to standard business intelligence functionalities. They focus on natural language descriptions of business data to highlight critical information. Quicksight automatically analyzes the created visualizations and suggests a number of textual narratives based off of them. The suggested insights are ready out-of-the box, but can also be customized to every need.
Quicksight Q – Amazon Quicksight Q uses natural language processing to provide users with answers to questions about their data. Q users do not need to familiarize themselves with the new tools, nor do they need to have technical knowledge in order to extract useful information from the data. Q administrators create datasets on the basis of which machine learning and natural language processing technology provide end users with answers to text input questions. Thus, for example, users can find out information about the best-selling goods in the desired period by customers by entering a query in the search engine.
Add a little bit of SPICE
Quicksight SPICE (Super-fast, Parallel, In-memory Calculation Engine) is a technology designed for efficient retrieval of analytical data with high performance. It represents a layer that stores data in-memory, which significantly speeds up the data retrieval process compared to the standard approach of sending direct queries to the databases where the data is located. SPICE uses in-memory technology, columnar data storage systems and machine code generation to achieve high data retrieval performance. In addition, SPICE also replicates data, which allows a large number of users to perform analysis simultaneously without losing performance. Because of that, there will be no reduction in performance in the form of slower reports, even if several tens of thousands of users simultaneously access the same report.
Quicksight and existing applications
Amazon offers a simple solution for incorporating reports, visualizations and dashboards created in Quicksight into existing applications or web portals. By using API functionality, or more simply, by copying the embed code, users can integrate their own applications with the robust Quicksight analytics functionalities to enable end users to perform advanced analytics operations on demand. Quicksight provides the ability to customize the dashboard by determining the amount of interactivity and by adding custom themes and selecting the elements that will be displayed in the application to create a unique user experience.
Quicksight’s serverless architecture enables customers to scale to a large number of users without the need to manage, maintain, upgrade or migrate their own servers. In addition, Quicksight allows connecting to a large number of data sources such as AWS services (Redshift, Athena, S3, RDS, Aurora…), third-party databases (Postgres, SQL Server, MySQL…), SaaS applications (Salesforce, JIRA, Github…), and different files (Excel, CSV, JSON…).