Workshops Use Cases Architecture Projects Blog
About Contact Offers Talks
Home›AI technologies›Text analysisDeutsch

Text analysis: structure out of text you already have

Natural language processing (NLP) reads text and returns structure: the entities in it, key phrases, sentiment, the language, and personal data that should not travel further.

Amazon ComprehendAzure AI LanguageNatural Language AIspaCy, or a model from Hugging Face
Book a free 30-min call →AI Concept and Prototype workshops →

What is text analysis?

Text analysis is usually the second step in a chain. Something else produced the text, a transcript or an OCR result, and this step decides what is in it: names, places, organisations, the tone, the topic.

Finding personal data (PII) is often the most useful part. It decides what may be stored, shown or sent on, and it can be the step that draws the line between cloud and building.

Goes inText, often the output of OCR or speech to text
Comes outEntities, key phrases, sentiment, language, personal data
Typical usesRedaction, routing tickets, tagging archives, compliance checks

Four ways to run it

Each of the big clouds offers it as a managed service, and you can also run it yourself. Same task, a different bill and a different console.

Amazon Comprehend
Managed on AWS

Amazon Comprehend[1]

The managed natural language processing service on AWS, including detection of personal data. You can train custom classifiers on your own labels.

Official documentation →
Azure AI Language
Managed on Microsoft Azure

Azure AI Language[2]

The managed language service on Microsoft Azure for entities, sentiment, personal data and summarisation.

Official documentation →
Natural Language AI
Managed on Google Cloud

Natural Language AI[3]

The managed text analysis API on Google Cloud.

Official documentation →
spaCy, or a model from Hugging Face
Run it yourself

spaCy, or a model from Hugging Face[4]

SpaCy is an open-source NLP library (MIT). For more specialised tasks, open models from Hugging Face run on your own machine.

Project page →
Amazon Comprehend: The managed service on one page, with the same service on the other platforms.
The managed service on one page, with the same service on the other platforms.[5]

How it looks in an architecture

Slides from my course on AI architectures at FH Burgenland, drawn from real projects.

Fifty-six questions sorted by what the data is: public, private or mixed. The classification decided where each one runs.
Fifty-six questions sorted by what the data is: public, private or mixed. The classification decided where each one runs.[5]
One question in the geospatial platform: two arrows cross the line, and both carry a question or an answer, never a record.
One question in the geospatial platform: two arrows cross the line, and both carry a question or an answer, never a record.[5]

Where I used it

The geospatial platform

In my geospatial platform, sorting data into public, private and mixed was the first step and it decided the architecture. Private parcel data never crosses to the cloud: agents on both sides exchange a question and an answer, and no names cross.

Architecture on ai-solutions.wiki →

Managed and self-hosted together: hybrid

With a ready-made service the call leaves your network. The hybrid decision is what crosses, and in what state: the whole file, a redacted one, or only the part that has to. Where a file may not leave at all, the same step runs with the open-source option inside your building, and only the result, or nothing, crosses.[6]

Hybrid is a decision at every layer. For text analysis the question is often which text may leave at all.
Hybrid is a decision at every layer. For text analysis the question is often which text may leave at all.[7]

What would help you next?

Learn to decide

The course “From Managed to Hybrid” teaches where AI should run: what each layer costs, what you operate and where your data has to stay. It is in preparation, with a free live session first.

Join the course waitlist →
Linda MohamedLinda Mohamed
AI and cloud consultant, AWS Community Hero, Vienna

Build it with me

Tell me what you want to build. In 30 minutes we check which of the four ways fits your data, your rules and your budget, and whether AWS funding for a proof of concept can cover part of it. How AWS funding works →

Book a free 30-min call →15-min question slot

You already know what you need

Book the workshop directly:

ai-solutions.wiki

Read more on ai-solutions.wiki

My open knowledge base, with longer comparisons, pricing notes and sources.

Sources

  1. Amazon Comprehend documentation. https://docs.aws.amazon.com/comprehend/latest/dg/what-is.html
  2. Azure AI Language documentation. https://learn.microsoft.com/en-us/azure/ai-services/language-service/overview
  3. Natural Language AI documentation. https://cloud.google.com/natural-language/docs
  4. spaCy, or a model from Hugging Face. https://spacy.io/
  5. Linda Mohamed, course slides “The projects”, MASE-3, FH Burgenland, September 2026
  6. NIST glossary: hybrid cloud. https://csrc.nist.gov/glossary/term/hybrid_cloud
  7. Linda Mohamed, course slides “Start here”, MASE-3, FH Burgenland, September 2026
  8. Amazon Comprehend, ai-solutions.wiki. https://ai-solutions.wiki/tools/amazon-comprehend/
  9. spaCy, ai-solutions.wiki. https://ai-solutions.wiki/tools/spacy/
  10. Hugging Face, ai-solutions.wiki. https://ai-solutions.wiki/tools/huggingface/
  11. GIS and AI architecture on AWS, ai-solutions.wiki. https://ai-solutions.wiki/solutions/geospatial/gis-ai-architecture/

Links checked September 2026. Product names and features change: check the linked pages before you decide.