Machine Learning on AWS
Amazon Web Services (AWS) offers a variety of tools and services for machine learning. These
include:








Amazon SageMaker: A fully managed service that enables developers and data
scientists to build, train, and deploy machine learning models quickly.
Tutorial: https://youtu.be/7QSsysGX14w
Amazon EC2 P3 Instances: High-performance, GPU-powered instances that are ideal
for training and deploying machine learning models.
Amazon Rekognition: A service that uses deep learning to recognize objects, scenes,
and faces in images and videos.
Amazon Lex: A service for building chatbots and voice assistants that can understand
and respond to natural language input.
Amazon Textract: A service that uses machine learning to extract text and data from
scanned documents.
Amazon Personalize: A service to create personalized recommendations for
customers. Once the model is trained, you can use the Amazon Personalize API to
generate personalized recommendations for individual customers in real-time.
Amazon Kendra: can index and search a variety of data sources, including documents,
websites, and databases. The service automatically extracts relevant information from
the data sources and stores it in a search index
Amazon Forecast: A service that uses machine learning to generate forecasts based
on time-series data.
Tutorial: https://youtu.be/0vR8_SkWJHI
Amazon Rekognition
• Find objects, people, text, scenes in images and videos using ML
• Facial analysis and facial search to do user verification, people counting
• Create a database of “familiar faces” or compare against celebrities
• Use cases:
o Labeling
o Content Moderation
o Text Detection
o Face Detection and Analysis (gender, age range, emotions...)
o Face Search and Verification
o Celebrity Recognition
o Pathing (ex: for sports game analysis)
Amazon Transcribe
• Automatically convert speech to text
• Uses a deep learning process called automatic speech recognition (ASR) to convert speech
to text quickly and accurately
• Automatically remove Personally Identifiable Information (PII) using Redaction
• Supports Automatic Language Identification for multi-lingual audio
• Use cases:
o transcribe customer service calls
o automate closed captioning and subtitling
o generate metadata for media assets to create a fully searchable archive
Amazon Polly
• Turn text into speech using deep learning
• Allowing you to create applications that talk
Amazon Translate
• Natural and accurate language translation
• Amazon Translate allows you to localize content - such as websites and applications - for
international users, and to easily translate large volumes of text efficiently.
Amazon Comprehend
• For Natural Language Processing – NLP
• Fully managed and serverless service
• Uses machine learning to find insights and relationships in text • Language of the text
o Extracts key phrases, places, people, brands, or events
o Understands how positive or negative the text is
o Analyzes text using tokenization and parts of speech
o Automatically organizes a collection of text files by topic
• Sample use cases:
o analyze customer interactions (emails) to find what leads to a positive or
negative experience
o Create and groups articles by topics that Comprehend will uncover