— the next-generation multilingual knowledge graph based on BabelNet

demo available soon

A huge multilingual knowledge graph

Knowledge graphs are the 21st century counterpart of dictionaries in previous centuries. They organize knowledge into a coherent network of meanings and they enable Artificial Intelligence applications which exploit this knowledge to perform text understanding.

WordAtlas is the next-generation knowledge graph. It greatly enhances BabelNet®, the award-winning multilingual semantic network, thanks to the know-how of years of research in computational linguistics in Prof. Roberto Navigli’s lab at the Sapienza University of Rome.

What makes WordAtlas special is its linkage between concepts and their surface realizations in hundreds of languages: WordAtlas provides millions of lexicalizations for each language, from common nouns, adjectives, verbs and adverbs, to hundreds of thousands of technical terms and millions of named entities, such as people, locations, organizations and products.

Each meaning comes with a wealth of information, including:

  • synonyms in dozens of languages
  • definitions (often multiple) in each language
  • taxonomical information: generalizations (i.e., more general concepts) and specializations (i.e. more specific concepts)
  • thousands of different relations (such as, part-of, used-for, born-in, located-in, related-to)
  • pictures which illustrate concepts

Power your natural language applications with WordAtlas now!

The Next-Generation BabelNet®

WordAtlas greatly innovates the award-winning BabelNet®, the largest multilingual semantic network, encyclopedic dictionary and knowledge repository for multilingual applications, which covers 14 million concepts and named entities in 270+ languages. BabelNet received the 2015 META prize for the “groundbreaking work in overcoming the language barriers” and the Prominent Paper Award 2017 from the Artificial Intelligence Journal. WordAtlas builds upon this success and takes BabelNet to the next level, by improving the underlying lexical-semantic model and providing professionally-curated content.

Babelnet logo


While BabelNet is a merger of open knowledge resources such as Wikipedia, Wiktionary and WordNet, WordAtlas adds professionally-curated language content and additional features in order to set a level playing field across languages and enable multilingual Natural Language Processing with high accuracy.

Product comparison

While BabelNet® is oriented towards the research community, WordAtlas is a commercial product which aims for a high accuracy of its content. On the right we compare statistics and product information for the two knowledge resources to show the strong impact of our knowledge graph in multilingual Natural Language Processing. Specifically, our professional curation of common nouns, verbs and adjectives, together with improved validation of open content, enables effective text understanding on a large scale.

BabelNet 4 WordAtlas
Number of languages 284 284
Availability free for public research institutions commercial
Validation of open content - professional
Support - professional
Last update February 2018 continuous