Developer Center

Overview

With the Evri API, you can automatically, cost effectively and in a fully scalable manner: analyze text, get recommendations, discover relationships, mine facts and get popularity data.

Evri API Overview

Getting Started

To get up and running quickly, here are some great first steps:

Analyze text with the Evri API

Identify, categorize and locate entities in text. Extract and graph how entities are interconnected. Automatically categorize documents. Extract contextual information for subsequent media recommendations so article, image, and video recommendations are accurately tailored to the topic of a trigger document.

Identify and locate entities

Identify people, places, and things that are present in unstructured text such as news articles, blog posts, Twitter tweets and other web content. Entities are recognized and assigned into one or more categories from our set of thousands including categories like: politician, actor, and artist. In addition to being assigned a category, returned entities have a unique identifier which allows you to look up exhaustive and complete factual information about the entity, such as birth date, films directed, songs sung, and more. Highly ambiguous entities such as Blue, Blue or Blue are accurately recognized and linked to detailed information about the specific Blue referenced in the text.

Entity identification

Entities can also be identified in very difficult language like Twitter tweets, Facebook status updates, or other terse colloquial use of language. In addition, entities can be easily located within submitted text.

Entity identification

Extract entity networks

Graph how entities within a news article, blog post, or other selection of text interrelate. Leverage these graphical relationships to provide additional, contextually tailored article, image, video, and tweet recommendations.

Entity networks

Categorize documents

Automatically categorize your documents into high level categories like politics, sports, and entertainment.

To get started analyzing text, see the REST spec section Get entity network about some text. Also, join our Developer Forum and contact us for licensing or partnership questions at api-licensing@evri.com.

Get recommendations with the Evri API

Get recommendations for news articles, blog posts, images, videos, Twitter tweets, quotes, sentiment and more. Get recommendations based on the context of an article or text snippet a user is reading. Get the most relevant articles, images, videos, or tweets to a particular politician, actor, film, sports team, or other entity even if the entity has a very ambiguous name like the album Ten.

Contextual media recommendations

Get recommendations based on the context of the news article, blog post or text snippet a user is reading. Get contextual recommendations with a focus on a particular entity mentioned in the article. For example, if Barack Obama is mentioned in an article about health care, get health care videos focused on Barack Obama and this particular article, i.e. not Barack Obama in the context of Afghanistan, the economy, or some other topic. Get contextual recommendations for a particular pair of entities to find how they are related. For example, get articles focused on Microsoft and Yahoo in the context of a particular trigger article.

Entity identification

Entity and graph driven recommendations

Get recommendations based on a person, place or thing in general, i.e. not in the context of a particular article. Given two entities, get recommended media about how they are related. For example, given the two entities Boeing 747 and United Airlines, find images about the two.

Entity networks

To get started getting recommendations, see the REST spec section Get media about an entity. The following example scenarios from the REST specification will help as well: Get recommended articles based on text, Get recommended videos for an entity based on text, and Get recommended articles for an entity. Also, join our Developer Forum and contact us for licensing or partnership questions at api-licensing@evri.com.

Discover relationships with the Evri API

Discover how your user’s favorite actors, products, or sports organizations are related. Discover new emergent relationships that link entities to an entity right now based on the breaking news, blog and other web content.

Action relationships

Discover relationships based on an action a person, place or thing is performing, or having performed against them. For example, find out who is criticizing Barack Obama right now, discover who is being attacked by a shark, or what actors are entering rehab.

Action relationships

Sentiment relationships

Discover relationships based on the sentiment a person, place or thing is expressing, or having expressed toward them. Specifically:

For a more in depth discussion of the Sentiment API, see the blog post: Sentiment API Exposes Web’s Feelings

Sentiment relationships

To get started discovering relationships, see the REST spec section Get relations about an entity, Get a related entity network about an entity, and Get sentiment information. Also, join our Developer Forum and contact us for licensing or partnership questions at api-licensing@evri.com.

Mine facts with the Evri API

Mine our knowledge base of facts involving millions of people, places, and things. We are busy aggregating and normalizing facts from many information repositories such as Wikipedia, Amazon, Crunchbase, Freebase and more. Find highly structured factual information about an entity, such as the day a person was born, the location of a business, or the members in a band. In addition, find out the common alias forms of a particular entity as mentioned in the press, such as Barack Obama, Obama, President Obama, Barack Hussein Obama, etc. Some sample uses include:

Facts

To get started mining facts, see the REST spec section Get information about an entity and Get entities and factual information. Also, join our Developer Forum and contact us for licensing or partnership questions at api-licensing@evri.com.

Get popularity data with the Evri API

Access popularity data including the most popular entities broken down by categories like politician, actor, sports person, disease, scientist, etc. Popularity is determined by frequency of mentions in news articles, blog posts, and other web content in addition to user activity data. Discover sudden spikes in popularity for a particular entity, then constrain recommended article searches to that day to find out why a person, place or thing suddenly became so popular.

Facts

To get started getting popularity data, see the REST spec section Get zeitgeist information. Also, join our Developer Forum and contact us for licensing or partnership questions at api-licensing@evri.com.

The Fine Print

Before proceeding, please review our Terms of Service. By accessing or using the Evri API, you are agreeing to be bound by that agreement. If you wish to reproduce or redistribute Evri’s documentation in whole or in part, you must retain this notice and a link to the agreement.

The Evri API is currently in an early preview stage. Only a small portion of our extensive knowledgebase and NLP based text analysis and search infrastructure is currently exposed. We are committed to exposing more and more of our powerful entity web platform over time; this preview release is the first step!

In addition, we are currently not requiring application ids or a sign up. But we will be adding these requirements soon, and will provide a time window for which you can transition to using an application id for your application.

Send Feedback!

We welcome any API feedback including ways to make the Evri API easier to use. Please use our Developer Forum to tell us about your successes and challenges. We will respond to your questions quickly.

You may also contact us at api-licensing@evri.com.