You're reading from Modern Data Architecture on AWS

Product typeBook

Published inAug 2023

PublisherPackt

ISBN-139781801813396

Edition1st Edition

Concepts

Data Science

Author (1)

Behram Irani

Predictive Analytics

A few years back, any discussion on artificial intelligence/machine learning (AI/ML) used to be a niche topic, relegated to the end chapters of most data platform books. The primary reason for this lack of urgency was due to the fact that AI/ML projects didn’t give a positive return on investment (ROI) for most businesses, due to the high total cost of ownership (TCO), for making AI/ML-based predictions a reality. However, with the onset of cloud technologies and the benefits it brings to businesses, AI/ML has become one of the primary topics of discussion and implementation for almost all businesses. We are now at a stage where any organization not doing any kind of predictive analytics is at risk of losing out to its competitors, who are constantly striving to look into the future and make business decisions based on it.

The topic of AI/ML is dense, and often, you will see a series of books catering to specific areas of it. Since we only have a chapter...

Role of AI/ML in predictive analytics

Before we get into the role of AI/ML, let’s quickly understand how AI, ML, and deep learning (DL) are co-related.

AI refers to a field of computer science that focuses on creating intelligent machines or systems that can perform tasks that typically require human intelligence. AI aims to simulate human cognitive processes such as learning, reasoning, problem-solving, perception, and language understanding. Out of the many possibilities, some examples of AI are speech recognition, computer vision (CV), natural language processing (NLP), learning, and problem-solving.

ML is a subfield of AI that focuses on the development of algorithms and models that allow computers to learn and make predictions or decisions without being explicitly programmed. ML also gets referred to as predictive analytics since it’s able to predict outcomes. Examples of ML usage in business terms would be sales forecasting, fraud detection, sentiment analysis...

Barriers to AI/ML adoption

For many years, AI/ML technology adoption was challenging for many organizations for many reasons. Let me quickly summarize some of them here:

...

AWS AI/ML services overview

AWS provides a very broad set of AI/ML services, right from specialized infrastructure and ML frameworks that allow ML engineers to custom train their ML models and deploy them on custom hardware. This includes ML frameworks such as PyTorch, TensorFlow, and Apache MXNet. ML infrastructure often requires plenty of CPU and GPU power. AWS provides many types of Amazon Elastic Compute Cloud (Amazon EC2) instances such as the P3 and Trn1 instances that are suitable for ML training. AWS also provides ML accelerators such as AWS Trainium for DL training and AWS Inferentia for high-performance ML inferences.

The next layer of services revolves around ML. AWS ML services are created specifically keeping in mind many of the barriers to ML adoption. In order to democratize ML, it is essential to have different services geared toward different personas in the organization. For the same reason, AWS has created ML services that can help multiple personas, even those...

AWS AI services, along with use cases

AWS offers over 20 AI services providing different capabilities in core as well as specialized categories. Core areas include vision, speech, text, and chatbots, whereas specialized areas include business processes, search, healthcare, industrial, DevOps, and generative AI. For the purpose of this chapter, we will not be able to cover each and every AWS AI service along with its use cases, but we will quickly summarize them here so that you are aware of when to use them for your specific use cases:

Challenge	Reasons
Expensive infrastructure	Training ML models on large datasets required a lot of compute, memory, and storage. Multiple iterations of tuning made this whole process very expensive on traditional on-prem infrastructure as all this hardware had to be procured upfront.
Not enough data scientists and ML builders	Building ML systems required niche skill sets with an understanding of complex ML algorithms. This made it difficult for organizations to easily acquire resources that had all the necessary skill sets to help them build an ML platform.
Tedious and time-consuming processes

ML using Amazon SageMaker, along with use cases

One of the biggest barriers to ML adoption has been that not everyone in the organization understands how the ML process works or has the skill sets to build an end-to-end ML platform. Amazon SageMaker is a comprehensive ML service that helps different personas easily use the platform to build, train, and deploy ML models for any use case. Data scientists want to quickly prepare the data to train and build ML models. ML engineers want to quickly deploy and manage these models at scale. Business analysts want to make ML predictions without having to learn ML technologies. This is where Amazon SageMaker as an ML platform helps. It’s a collection of tools that make every step of the ML process easier, faster, and cheaper to implement for different personas in the organization. The following diagram depicts this aspect of SageMaker:

Figure 10.6 – Amazon SageMaker user personas

Let’s get...

ML using Amazon Redshift and Amazon Athena

Many times, all the data is already processed, stored, and consumed out of Amazon Redshift using SQL-based queries. Database engineers can easily create complex SQL-based consumption patterns, but they lack the understanding to stitch together all the components of ML pipelines using SageMaker. To make their day-to-day-job lives easy, they can now build ML models inside Amazon Redshift using SQL syntax. Redshift ML handles all interactions with Amazon SageMaker, transparent to the data developer.

Some of the benefits of using Redshift ML are set out here:

Simplicity: Makes it easy to create ML models using SQL. Even the predictions are done using SQL statements.
Flexibility: Allows the user to select specific ML algorithms such as XGBoost. Under the covers, the best ML model is automatically trained and tuned.
Performant: Even though under the covers the models are trained with SageMaker, they are eventually deployed in...

Summary

In this chapter, we looked at how AI/ML technologies play a big role in predictive analytics so that organizations can stay ahead of the curve and proactively make decisions before things happen. But at the same time, we also looked at many of the barriers related to the adoption of AI/ML and how AWS is able to overcome all these barriers.

We introduced the different stacks of how AWS provides services specific to each of these layers. For the AI layer, AWS provides a long list of 20+ services that help with specific types of AI problems such as speech, image, text, and so forth. These services help fast-track solutions that can be solved by pre-trained ML models.

We then looked at Amazon SageMaker as an ML service that has many components to it. SageMaker Canvas helps business analysts with low-code/no-code types of tools so that they can quickly create ML models and predict business outcomes. We looked at how SageMaker Studio has various tools inside it to help with...

References

Amazon SageMaker Canvas workshop: https://catalog.us-east-1.prod.workshops.aws/workshops/80ba0ea5-7cf9-4b8c-9d3f-1cd988b6c071/en-US
Amazon SageMaker workshop: https://catalog.us-east-1.prod.workshops.aws/workshops/63069e26-921c-4ce1-9cc7-dd882ff62575/en-US
Amazon SageMaker Data Wrangler workshop: https://catalog.us-east-1.prod.workshops.aws/workshops/327375b8-425b-49d4-b0da-241da0595ecf/en-US
Amazon Redshift ML workshop: https://catalog.us-east-1.prod.workshops.aws/workshops/4efa7f96-66a8-4b39-b7ea-c34595b2352b/en-US
Content moderation workshop using AWS AI services: https://catalog.us-east-1.prod.workshops.aws/workshops/1ece9ffd-4c24-4e66-b42a-0c0e13b0f668/en-US

The rest of the chapter is locked

You have been reading a chapter from

Modern Data Architecture on AWS

Published in: Aug 2023Publisher: PacktISBN-13: 9781801813396

A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.

undefined

Unlock this book and the full library FREE for 7 days

Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of

Start free trial

Renews at $15.99/month. Cancel anytime

Author (1)

Behram Irani

Behram Irani is currently a technology leader with Amazon Web Services (AWS) specializing in data, analytics and AI/ML. He has spent over 18 years in the tech industry helping organizations, from start-ups to large-scale enterprises, modernize their data platforms. In the last 6 years working at AWS, Behram has been a thought leader in the data, analytics and AI/ML space; publishing multiple papers and leading the digital transformation efforts for many organizations across the globe. Behram has completed his Bachelor of Engineering in Computer Science from the University of Pune and has an MBA degree from the University of Florida.
Read more about Behram Irani

Personalised recommendations for you

Based on your interests and search pattern

Et al.

Ever wonder why speech recognition systems don't understand the Scottish accent, or what would happen if an astronaut only ate mac 'n' cheese, or other spurious reflections you'd have at a bar? We did, then collated those deliberations into absurd research articles with fake figures and methodologies inspired by even more fictionally absurd studies.

BookAug 2023230 pages5

Generative AI with LangChain

This book is a comprehensive introduction to LLMs and LangChain, demystifying the basic mechanics of LangChain, its functionalities, and the myriad of applications it can be integrated into.

BookDec 2023360 pages4

Generative AI with LangChain

This book is a comprehensive introduction to LLMs and LangChain, demystifying the basic mechanics of LangChain, its functionalities, and the myriad of applications it can be integrated into.

BookDec 2023360 pages5

Generative AI with LangChain

This book is a comprehensive introduction to LLMs and LangChain, demystifying the basic mechanics of LangChain, its functionalities, and the myriad of applications it can be integrated into.

BookDec 2023360 pages1

Generative AI with LangChain

This book is a comprehensive introduction to LLMs and LangChain, demystifying the basic mechanics of LangChain, its functionalities, and the myriad of applications it can be integrated into.

BookDec 2023360 pages5

Mastering Tableau 2023

This book is a comprehensive resource to mastering your Tableau skills and becoming a BI expert. As you progress, you will learn how to build advanced dashboards and improve your storytelling to derive key business insight, as well as make you well-versed with advanced functionalities of Tableau in the business intelligence domain.

BookAug 2023684 pages

Building AI Applications with ChatGPT APIs

This guide covers all ChatGPT API features for effortless creation of robust AI powered apps. With its help, you’ll be able to leverage ChatGPT’s cutting-edge NLP models to take your app development skills to the next level. You’ll also work on ten exciting projects that will give you the practical know-how that you can apply to your existing applications.

BookSep 2023258 pages5

Building AI Applications with ChatGPT APIs

This guide covers all ChatGPT API features for effortless creation of robust AI powered apps. With its help, you’ll be able to leverage ChatGPT’s cutting-edge NLP models to take your app development skills to the next level. You’ll also work on ten exciting projects that will give you the practical know-how that you can apply to your existing applications.

BookSep 2023258 pages2

Data Engineering with AWS

Embark on a journey to master data engineering pipelines on AWS! Our book offers a hands-on experience of AWS services for ingesting, transforming, and consuming data. Whether you're an absolute beginner or someone with basic data engineering experience, this guide is an indispensable resource.

BookOct 2023636 pages5

Modern Data Architecture on AWS

Every organization wants an agile, performant, and cost-effective data platform that meets all their current and future business needs. Purpose-built AWS analytics services and their features play a big part in building such a modern data platform. This book brings to you all the design and architectural patterns that’ll help you achieve this goal.

BookAug 2023420 pages5

Practical Guide to Applied Conformal Prediction in Python

Discover the power of Conformal Prediction with the "Practical Guide to Applied Conformal Prediction in Python." Master the latest techniques to quantify uncertainty in machine learning and computer vision models, and seamlessly apply them to your industry applications.

BookDec 2023240 pages

TinyML Cookbook

With over 70 project-based recipes, the TinyML Cookbook is a practical guide that will help you to get the most out of your microcontrollers. It provides a comprehensive understanding of the theoretical foundations while giving you hands-on experience training ML models for deployment on Arduino Nano 33 BLE Sense, Raspberry Pi Pico, and SparkFun RedBoard Artemis Nano microcontrollers.

BookNov 2023664 pages

AWS AI service

Description

Common use cases

Amazon Rekognition

Makes it easy to perform image as well as video analysis. You can build apps that leverage Rekognition to identify people, text, objects, and other activities from pictures and videos.

Identity verification...