Skip to content

AWS Machine Learning and AI services

Intro

AWS Analytics services

Analytics

Description

Sure, here are the Machine Learning and AI services provided by AWS with direct markdown links:

  • Amazon Rekognition: Analyzes images and videos for cataloging assets, automating workflows, and extracting meaning.
  • Amazon Lookout for Vision: Detects defects and automates inspection for comprehensive quality control.
  • AWS Panorama: Utilizes computer vision at the edge to improve operations with automated monitoring.
  • Amazon Textract: Extracts text and data from millions of documents at speed.
  • Amazon Comprehend: Maximizes the value of unstructured text with natural language processing (NLP).
  • Amazon A2I: Adds humans to the review process to ensure accuracy and compliance of sensitive data.
  • Amazon Lex: Builds chatbots & virtual agents for automated conversation channels.
  • Amazon Transcribe: Enhances applications and workflows with automatic speech recognition.
  • Amazon Polly: Converts text into life-like speech.
  • Amazon Kendra: Search queries in natural language, like "where can I get tested for COVID19", usable on websites and applications.
  • Amazon Personalize: Uses ML to customize applications and websites to each individual user. https://www.youtube.com/watch?v=y7crGW6zLao
  • Amazon Translate: Expands your reach and accessibility with fast, accurate, and customizable translation.
  • Amazon Forecast: Harnesses unique data types and time series data to create accurate end-to-end prediction models.
  • Amazon Fraud Detector: Stops adversaries and identifies potential attacks.
  • Amazon Lookout for Metrics: Detects and identifies root causes of unexpected changes in metrics.
  • Amazon DevOps Guru: Simplifies operational performance measurement and reduces application downtime.
  • Amazon SageMaker: It’s a fully managed service that enables data scientists and developers to quickly and easily build, train, and deploy machine learning models at any scale

These services are designed to provide ready-made intelligence for your applications and workflows, and they easily integrate with your applications to address common use cases. They are continuously learning and improving, providing you with quality and accuracy.

AWS Fraud detection

Part A

  • Build,
  • train, and
  • deploy

an Amazon Fraud Detector model.

Reference: https://docs.aws.amazon.com/frauddetector/latest/ug/part-a.html

Part B

Generate fraud predictions.

The steps :

  • Build Detector
  • Add model
  • Add rules e.g. inference>=80 = high_fraud_probability, 80>inference && inference>20 = medium_fraud_probability
  • Configure rule execution and rule order
  • Review and create detector version
  • Test and activate detector version

Reference: https://docs.aws.amazon.com/frauddetector/latest/ug/part-b.html

Example: https://aws.amazon.com/blogs/machine-learning/prevent-fake-account-sign-ups-in-real-time-with-ai-using-amazon-fraud-detector/