About
With IBM Watson® Natural Language Understanding, developers can analyze semantic features of text input, including categories, concepts, emotion, entities, keywords, metadata, relations, semantic roles, and sentiment.
Features
Send requests to the API with text, HTML, or a public URL, and specify one or more of the following features to analyze:
Categories
Categorize your content using a five-level classification hierarchy. View the complete list of categories here. For example:
Input
url: "www.cnn.com"
Response
/news
/art and entertainment
/movies and tv/television
/news
/international news
Concepts
Identify high-level concepts that aren't necessarily directly referenced in the text. For example:
Input
text: "Natural Language Understanding uses natural language processing to analyze text."
Response
Linguistics
Natural language processing
Natural language understanding
Emotion
Analyze emotion conveyed by specific target phrases or by the document as a whole. You can also enable emotion analysis for entities and keywords that are automatically detected by the service. For example:
Input
text: "I love apples, but I hate oranges."
targets: "apples", and "oranges"
Response
"apples": joy
"oranges": anger
Entities
Find people, places, events, and other types of entities mentioned in your content. View the complete list of entity types and subtypes here. For example:
Input
text: "IBM is an American multinational technology company headquartered in Armonk, New York, United States, with operations in over 170 countries."
Response
IBM: Company
Armonk: Location
New York: Location
United States: Location
Keywords
Search your content for relevant keywords. For example:
Input
url: "http://www-03.ibm.com/press/us/en/pressrelease/51493.wss"
Response
Australian Open
Tennis Australia
IBM SlamTracker analytics
Metadata
For HTML and URL input, get the author of the webpage, the page title, and the publication date. For example:
Input
url: "https://www.ibm.com/blogs/think/2017/01/cognitive-grid/"
Response
Author: Stephen Callahan
Title: Girding the Grid with Cognitive Computing - THINK Blog
Publication date: January 31, 2017
Relations
Recognize when two entities are related, and identify the type of relation. For example:
Input
text: "The Nobel Prize in Physics 1921 was awarded to Albert Einstein."
Response
"awardedTo" relation between "Noble Prize in Physics" and "Albert Einstein"
"timeOf" relation between "1921" and "awarded"
Semantic Roles
Parse sentences into subject-action-object form, and identify entities and keywords that are subjects or objects of an action. For example:
Input
text: "In 2011, Watson competed on Jeopardy!"
Response
Subject: Watson
Action: competed
Object: on Jeopardy
Sentiment
Analyze the sentiment toward specific target phrases and the sentiment of the document as a whole. You can also get sentiment information for detected entities and keywords by enabling the sentiment option for those features. For example:
Input
text: "Thank you and have a nice day!"
Response
Positive sentiment (score: 0.91)
Syntax
Identify the sentences and tokens in your text. For example:
Input
text: "I love apples! I do not like oranges."
Response
| Sentence | Location |
|---|---|
| "I love apples!" | [0, 14] |
| "I do not like oranges." | [15,37] |
| Token | Lemma | Part of Speech | Location |
|---|---|---|---|
| "I" | "I" | PRON |
[0, 1] |
| "love" | "love" | VERB |
[2, 6] |
| "apples" | "apple" | NOUN |
[7, 13] |
| "!" | PUNCT |
[13, 14] |
|
| "I" | "I" | PRON |
[15, 16] |
| "do" | "do" | AUX |
[17, 19] |
| "not" | "not" | PART |
[20, 23] |
| "like" | "like" | VERB |
[24, 28] |
| "oranges" | "orange" | NOUN |
[29, 36] |
| "." | NOUN |
[36, 37] |
Supported languages
See the Language support documentation for details about supported languages in Natural Language Understanding.