Azure Cognitive Services provides different predefined AI capabilities, which are categorized depending upon their capability, such as Vision, Speech, Language and Decision.
Communication with computers conventionally happens by giving commands such as through clicking buttons or pressing a key. Natural Language Processing is a form of Artificial Intelligence (AI) which makes the computers understand human languages such as English. This has significant advantage as now you can converse with computers rather than giving specific set of commands.
For example, in a chatbot application, you can type text and expect the application to return the response in English. Another example is when a user writes a bad review about a product and the application is able to recognize it as a negative sentiment. This analysis of large amount of unstructured text data can be done much faster than computers than by humans.
Not only this using machine learning NLP can learn about the conext and meaning of words and phrases.
Azure NLP is part of a category of services called Azure Cognitive Service for Language.
Cognitive service for language includes three services Text Analytics, QnA Maker, and Language Understanding or LUIS.
Text Analytics
This provides the capability to perform different tasks on the given text such as,
- Key phrase extraction
Identifying the main text elements in the text.
- Named Entity Recognition (NER)
Identifying the main entities such as people and places.
- Language detection
Identifying Language of the text such as English or German.
- Sentiment analysis
Identifying the text shows positive, negative or neutral sentiments.
QnA Maker
For defining answers to questions from a knowledge base.
You can create this Knowledge base or KB from different sources such as pdf files, HTML documents, and word documents.
You create this KB by importing a different set of documents that contains information in questions and answers format such as FAQs(frequently asked questions)
This KB can then be used for tasks such as responding to users' questions with the appropriate answers.
Language Understanding or LUIS
This is used for extracting meaning from text by using machine learning models.
LUIS is accessed through its own portal https://www.luis.ai/
![]()
Another option available now is Azure OpenAI https://azure.microsoft.com/en-us/products/ai-services/openai-service