15. ML – Other AWS Services – Yannick's non-Bakery

Amazon Comperehend

Natural Language Processing and Text Analytics
Input social media, emails, web pages, documents, transcripts, medical records (Comprehend Medical)
Extract key phrases, entities, sentiment, language, syntax, topics, and document classifications
Events detection
PII Identification & Redaction
Targeted sentiment (for specific entities)
Compose
- Entities: Noun with Category, Confidence
- Key phrases: Noun
- Language
- Sentiment: Neutral, Positive, Negative, Mixed
- Syntax: Noun, Verb, Adposition, Adjustive, …

Amazon Translate

Uses deep learning for translation
Supports custom terminology
In CSV or TMX format
Appropriate for proper names, brand names, etc.

Amazon Transcribe

Speech to text
- Input in FLAC, MP3, MP4, or WAV, in a specified language
- Streaming audio supported (HTTP/2 or WebSocket)
Speaker Identificiation
- Specify number of speakers
Channel Identification
- i.e., two callers could be transcribed separately
- Merging based on timing of “utterances”
Automatic Language Identification
Custom Vocabularies
- Vocabulary Lists (just a list of special words – names, acronyms)
- Vocabulary Tables (can include “SoundsLike”, “IPA”, and “DisplayAs”)
Practices
- Call Analytics
- Medical
- Subtitling

Amazon Polly

Neural Text-To-Speech
Lexicons
- Customize pronunciation of specific words & phrases
- Example: “World Wide Web Consortium” instead of “W3C”
SSML
- Speech Synthesis Markup Language
- Gives control over emphasis, pronunciation, breathing, whispering, speech rate, pitch, pauses.
Speech Marks
- metadata can encode when sentence / word starts and ends in the audio stream
- Useful for lip-synching animation

Amazon Rekognition

Computer vision
Object and scene detection
Image moderation
Facial analysis
Celebrity recognition
Face comparison
Text in image
Video analysis
- Objects / people / celebrities marked on timeline
- People Pathing
The Nitty Gritty
- Images come from S3, or provide image bytes as part of request
- Facial recognition depends on good lighting, angle, visibility of eyes, resolution
- Video must come from Kinesis Video Streams
  - H.264 encoded
  - 5-30 FPS
  - Favor resolution over framerate
- Can use with Lambda to trigger image analysis upon upload
- Can use Custom Labels

Amazon Forecast

“AutoML” chooses best model for your time series data
- CNN-QR
  - Convolutional Neural Network – Quantile Regression
  - Best for large datasets with hundreds of time series
  - Accepts related historical time series data & metadata
- DeepAR+
  - Recurrent Neural Network
  - Best for large datasets
  - Accepts related forward-looking time series & metadata
- Prophet
  - Additive model with non-linear trends and seasonality
- NPTS
  - Non-Parametric Time Series
  - Good for sparse data. Has variants for seasonal / climatological forecasts
- ARIMA
  - Autoregressive Integrated Moving Average
  - Commonly used for simple datasets (<100 time series)
- ETS
  - Exponential Smoothing
  - Commonly used for simple datasets (<100 time series)
Works with any time series
- Price, promotions, economic performance, etc.
- Can combine with associated data to find relationships
Inventory planning, financial planning, resource planning
Based on “dataset groups,” “predictors,” and “forecasts.”

Amazon Lex

Natural-language chatbot engine
A Bot is built around Intents
- Utterances invoke intents (“I want to order a pizza”)
- Lambda functions are invoked to fulfill the intent
- Slots specify extra information needed by the intent
  - Pizza size, toppings, crust type, when to deliver, etc.
Can deploy to AWS Mobile SDK, Facebook Messenger, Slack, and Twilio
Lex Automated Chatbot Designer
- provide existing conversation transcripts
- Lex applies NLP & deep learning, removing overlaps & ambiguity
- Intents, user requests, phrases, values for slots are extracted
- Ensures intents are well defined and separated
- Integrates with Amazon Connect transcripts

Amazon Personalise

Fully-managed recommender engine
API access
- Feed in data (purchases, ratings, impressions, cart adds, catalog, user demographics etc.) via S3 or API integration
- provide an explicit schema in Avro format
- Javascript or SDK
- GetRecommendations
  - Recommended products, content, etc.
  - Similar items
- GetPersonalizedRanking
  - Rank a list of items provided
  - Allows editorial control / curation
Features
- Real-time or batch recommendations
- Recommendations for new users and new items (the cold start
  problem)
- Contextual recommendations
  - Device type, time, etc.
- Similar items
- Unstructured text input
- Intelligent user segmentation
  - For marketing campaigns
- Business rules and filters
- Promotions
  - Inject promoted content into recommendations
  - Can find most relevant promoted content
- Trending Now
- Personalized Rankings
Terminology
- Datasets
  - Users, Items, Interactions
- Recipes
  - USER_PERSONALIZATION
  - PERSONALIZED_RANKING
  - RELATED_ITEMS
  - USER_SEGMENTATION
- Solutions
  - Trains the model
  - Optimizes for relevance as well as your additional objectives
    - Video length, price, etc. – must be numeric
  - Hyperparameter Optimization (HPO)
- Campaigns
  - Deploys your “solution version”
  - Deploys capacity for generating real-time
    recommendations
Hyperparameter
- User-Personalization, Personalized-Ranking
  - hidden_dimension (HPO)
  - bptt (back-propagation through time – RNN)
  - recency_mask (weights recent events)
  - min/max_user_history_length_percentile (filter out robots)
  - exploration_weight 0-1, controls relevance
  - exploration_item_age_cut_off – how far back in time you go
- Similar-items
  - item_id_hidden_dim (HPO)
  - item_metadata_hidden_dim (HPO with min & max range specified)
Maintaining Relevance
- Use PutEvents operation to feed in real-time user behavior
- Retrain the model
  - They call this a new solution version
  - Updates every 2 hours by default
  - Should do a full retrain (trainingMode=FULL) weekly

Amazon TextTract

OCR with forms, fields, tables support

AWS DeepRacer

Reinforcement learning powered 1/18-scale race car

Amazon Lookout

Equipment, metrics, vision
Detects abnormalities from sensor data automatically to detect equipment issues
Monitors metrics from S3, RDS, Redshift, 3rd party SaaS apps
Vision uses computer vision to detect defects in silicon wafers, circuit boards, etc.

Amazon Monitron

End to end system for monitoring industrial equipment & predictive maintenance

TorchServe

Model serving framework for PyTorch
Part of the PyTorch open source project from Facebook (Meta?)

AWS Neuron

SDK for ML inference specifically on AWS Inferentia chips
EC2 Inf1 instance type
Integrated with SageMaker or whatever else you want (deep learning AMI’s, containers, Tensorflow, PyTorch, MXNet)

AWS Panorama

Computer Vision (SDK) at the edge
Brings computer vision to your existing IP cameras

AWS Deep Composer

AI-powered (music) keyboard
Composes a melody into an entire song
For educational purposes

Amazon Fraud Detection

Upload your own historical fraud data
Builds custom models from a template you choose
Exposes an API for your online application
Assess risk from:
- New accounts
- Guest checkout
- “Try before you buy” abuse
- Online payments

Amzon CodeGuru

Part of CICD
Automated code reviews!
Finds lines of code that hurt performance
Resource leaks, race conditions
Fix security vulnerabilities
Offers specific recommendations
Powered by ML
Supports Java and Python

Amazon Kendra

Enterprise search with natural language
Combines data from file systems, SharePoint, intranet, sharing services (JDBC, S3) into one searchable repository
ML-powered (of course) – uses thumbs up / down feedback
Relevance tuning – boost strength of document freshness, view counts, etc.

Use Cases	Amazon Kendra	Elasticsearch	OpenSearch
Improve search experiences	Yes	Yes	Yes
Enhance customer interactions and satisfaction	Yes	Yes	Yes
Natural Language Processing (NLP)	Yes	Yes, but with the help of additional configurations or integrations.	Yes, but it requires external NLP tools.
Full-text search	Yes	Yes	Yes,
Enterprise search	Yes	Yes, but it requires extra configuration.	Yes, but it needs extra configuration.
Distributed search	Yes	Yes	Yes
Piped query language search	No	Yes	Yes
Integrate search functionality into your SaaS applications	Yes	Yes	Yes
Implement multi-tenancy and get rid of fraud and risk	Yes	Yes	Yes
Semantic search	Yes	Yes	Yes
Application search	Yes	Yes	Yes
E-commerce search	No	Yes	Yes
Log and event data monitoring	No	Yes	Yes
Aggregate and analyze large datasets	No	Yes	Yes
Security Information and Event Management (SIEM)	No	Yes	Yes

Amazon Augmented AI (A2I)

Human review of ML predictions
Builds workflows for reviewing low-confidence predictions
Access the Mechanical Turk workforce or vendors
Integrated into Amazon Textract and Rekognition
Integrates with SageMaker
Difference with SageMaker Ground Truth
- Focus:
  - A2I: Focuses on integrating human review into the decision-making process of models post-prediction.
  - Ground Truth: Focuses on creating high-quality labeled datasets during the training phase.
- Stage of ML Workflow:
  - A2I: Used during the model deployment and prediction phase.
  - Ground Truth: Used during the data preparation and model training phase.
- Automation:
  - A2I: Adds human review when machine confidence is low.
  - Ground Truth: Combines automated labeling with human corrections to produce accurate training data.

Amazon Connect

an Amazon Web Services (AWS) public cloud customer contact center service
Contact Lens
- For customer support call centers
- Ingests audio data from recorded calls
- Allows search on calls / chats
- Sentiment analysis
- Find “utterances” that correlate with successful calls
- Categorize calls automatically
- Measure talk speed and interruptions
- Theme detection: discovers emerging issues