What is Amazon Comprehend?
Amazon Comprehend is a fully managed Natural Language Processing (NLP) service that uses machine learning to extract insights from unstructured text. The service analyzes documents, emails, support tickets, social media posts, and other text sources without requiring your own ML expertise.
Comprehend automatically detects the language of a text, identifies key entities (people, places, organizations, dates), analyzes sentiment (positive, negative, neutral, mixed), extracts key phrases, and detects personally identifiable information (PII). For healthcare, the separate Amazon Comprehend Medical offers specialized capabilities for extracting medical information.
Note: The former Comprehend features Topic Modeling, Event Detection, and Prompt Safety Classification were closed to new customers as of April 30, 2026; existing users with usage in the past 12 months retain access. For new projects in this area, AWS points to Amazon Bedrock (LLM-based topic detection) and Amazon Bedrock Guardrails (prompt safety).
Core Features
- Sentiment Analysis: Detects emotional sentiment in text at document and sentence level
- Entity Recognition: Identifies people, places, organizations, dates, quantities, and other entity types
- Key Phrase Extraction: Finds the most important terms and phrases in text
- Language Detection: Automatically detects a wide range of languages
- PII Detection and Redaction: Identifies and masks personally identifiable information in text
- Custom Classification and Custom Entity Recognition: Train your own models for specific categories and entity types
Typical Use Cases
Customer Feedback Analysis: Automatically analyze customer reviews, support tickets, and social media comments. Comprehend detects sentiment and extracts product names and problem areas.
Document Classification: Automatically sort incoming documents into categories such as invoices, contracts, complaints, or inquiries. With Custom Classification, train models for your specific document types.
Privacy and Compliance: Comprehend detects personally identifiable information (PII) in free text and can mask or redact it for compliance purposes, for example before further processing or archiving documents.
Benefits
- No ML knowledge required: Ready to use immediately via API
- Scales automatically from individual documents to large volumes of text
- Integrates seamlessly with S3, Lambda, and other AWS services
- Custom Models enable industry-specific customizations
Integration with innFactory
As an AWS Reseller, innFactory supports you with Amazon Comprehend: architecture, integration into existing systems, training of custom models, migrating individual use cases to Amazon Bedrock, and cost optimization.
Typical Use Cases
Frequently Asked Questions
What is Amazon Comprehend?
Amazon Comprehend is a fully managed NLP (Natural Language Processing) service that uses machine learning to extract insights from text. The service detects language, entities, key phrases, and sentiment in unstructured text.
Which languages does Amazon Comprehend support?
Amazon Comprehend supports German, English, Spanish, French, Italian, Portuguese, Japanese, Korean, Hindi, Arabic, and simplified and traditional Chinese, among others, for its core features. Feature coverage can vary by language.
What is Amazon Comprehend Medical?
Amazon Comprehend Medical is a separate, specialized variant for healthcare. The service extracts medical information such as diagnoses, medications, dosages, and treatments from clinical text.
Can I train custom models with Comprehend?
Yes, with Custom Classification and Custom Entity Recognition you can train your own classification and entity recognition models on your data to detect industry-specific terms and categories.
What happened to Topic Modeling and Event Detection in Comprehend?
As of April 2026, the Comprehend Topic Modeling, Event Detection, and Prompt Safety Classification features are no longer available to new customers, while existing users retain access. AWS recommends using Amazon Bedrock, and Bedrock Guardrails for prompt safety, for new projects in this area.
Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of AWS (official documentation). This page does not represent an offer by AWS.