- "Do I Know This Already?" Quiz
- Artificial Intelligence/Machine Learning Services
- Data Analytics Services
- Review All Key Topics
- Define Key Terms
- Q&A
This chapter is from the book
This chapter is from the book
This chapter is from the book
Artificial Intelligence/Machine Learning Services
I don’t want to take anything for granted in this section, so let’s begin by defining AI and ML. Artificial intelligence (AI) refers to computer systems or machines that are designed to perform tasks that typically require human intelligence, such as learning, reasoning, problem-solving, and decision-making. A subset of this exciting discipline is machine learning (ML), which involves the algorithms and models that enable computers to learn patterns from data and make predictions or decisions without explicit programming.
AI and ML are lofty disciplines that typically require the latest and greatest technologies and lots of available resources (like CPU, memory, and storage). AWS is perfectly positioned to help companies take advantage of these cutting-edge technologies.
SageMaker
AWS SageMaker is a smart assistant that you can use to build and train machine learning models without needing to be a coding expert. It provides easy-to-use tools to help you gather and prepare data, pick the right algorithm, and then train and deploy your model, all in one convenient place on the AWS Cloud platform. Figure 17-1 shows AWS SageMaker in the AWS Management Console.
)
Figure 17.1 AWS SageMaker
AWS SageMaker offers several features that simplify the ML lifecycle. Here are just some of them:
Built-in algorithms: SageMaker comes with a variety of prebuilt algorithms for common ML tasks—such as classification, regression, and clustering— which means you don’t need to create models from scratch.
Notebook instances: SageMaker provides Jupyter notebook instances, which allow you to create and share documents that contain live code, equations, visualizations, and narrative text.
Training jobs: You can use SageMaker to easily train your ML models at scale, distributing the training process across multiple instances.
Model hosting: Once your model is trained, SageMaker makes it simple to deploy, host, and integrate it with your applications.
Managed endpoints: SageMaker provides managed endpoints for deploying models, making it easy to handle real-time predictions and batch processing.
Autopilot: SageMaker enables you to automate the end-to-end process of building, training, and deploying ML models with minimal effort, making it suitable for users with limited ML expertise.
Lex
AWS Lex makes it easy to create virtual assistants for your applications. It’s a service that helps you build chatbots and conversational interfaces using natural language understanding. Think of it as the brain behind a chatbot. Lex understands user inputs, extracts key information, and can respond in a way that makes sense.
This service is handy for creating interactive experiences in your applications, whether for answering customer queries, handling reservations, or guiding users through processes. You can integrate Lex into various platforms, such as mobile apps or websites, to make it easier for users to interact with your applications using just their words. Plus, Lex is powered by the same technology as Amazon Alexa, so it’s got some serious language smarts under the hood.
Kendra
AWS Kendra is a super-smart search engine that is designed to help you find information effortlessly. It’s a powerful search service that uses ML to understand the context and meaning behind your queries. Instead of just matching keywords, Kendra comprehends natural language, making it feel like you’re having a conversation with your search engine. It’s great for handling complex searches across vast amounts of data in documents, FAQs, or other sources. Figure 17-2 shows the Kendra service in the AWS Management Console.
)
Figure 17.2 AWS Kendra
AWS Kendra includes the following features:
Semantic search: Kendra uses machine learning algorithms to understand the semantics of the content and improve the accuracy of search results by recognizing nuances of and relationships between words.
Relevance tuning: Kendra allows you to fine-tune search results to prioritize certain documents or sources based on your preferences. It enables you to ensure that the most important information is surfaced first.
Rich document support: Kendra can handle a variety of document types, including PDFs, Word documents, HTML, and more, making it versatile for different types of content.
Query suggestions: Kendra provides query suggestions to guide users and help them refine their search queries for better results.
Natural language query enhancement: Kendra assists users in constructing more effective queries by suggesting natural language improvements, making the search process more intuitive.