For Immediate Release: Sep 16, 2020

Diffbot Unveils New Product: Industry-First Knowledge as a Service Natural Language Processing System

At Diffbot, we believe that access to structured information will be the critical resource for the coming wave of intelligent applications—for everything from new app experiences, to search assistants, to enterprise business intelligence.

September 16, 2020 – Menlo Park, California – Diffbot, a leading Knowledge as a Service (KaaS) company powering intelligent web applications, today announced the launch of its Natural Language Processing (NLP) API, a new deep learning system that autonomously assesses information in free text via external knowledge on the web. The new NLP system is integrated into the company’s— the world’s largest—multilingual Knowledge Graph, which analyzes, extracts, and enhances content across the unstructured web, to learn and deconstruct natural language text.

In addition, the company achieved its second industry breakthrough in a decade using free web text to generate the most accurate character genealogy in two of the most popular biblical and cultural texts in human history. Diffbot’s NLP API was able to completely analyze the Bible and accurately generated a full genealogy of all the historical figures mentioned in the ancient text. The system also successfully evaluated George R. R. Martin’s epic fantasy novel series A Song of Ice and Fire (the basis for the hit HBO series Game of Thrones) and built a family tree of 3663 characters, larger than any such tree built by the human fans.

“With this new breakthrough, Diffbot continues to push the boundaries in achieving its vision in building an ImageNet for human language and realizing a Semantic Web that will transform the business world,” said Mike Tung, CEO of Diffbot. “We have been testing its capabilities in beta and the results have been extremely promising in illustrating the true impact it will have on the professional world, especially in the fields of science, history, education, and religion.” Tung continued: “I believe we are creating a system that will serve as a dedicated enterprise search engine—the real Semantic Web—unbound by the ad-support model where businesses in every industry can find accurate and extensive information and build customized workflows on top of their queries.”

Diffbot’s NLP API leverages machine learning, which has data models trained on words and names in every language via its Knowledge Graph, to verify the information and the inference it came from. It essentially functions as a structured knowledge index for any query or search, even rare manuscripts. By connecting documents to knowledge from around the web, Diffbot’s Natural Language Processing system will unlock the black box of information and pave the way for achieving general intelligence. And through fostering innovation on the web using intelligent data models such as artificial intelligence, machine learning, and computer vision, Diffbot has created a Knowledge as a Service ecosystem that is capable of providing fact-based information to any query.

The Natural Language Processing system is now available. To learn more about Diffbot, inquire about pricing, or book a demo, please visit

* Benchmarks are based on published papers from the above listed entities. Links to all pertinent citations available here:

About Diffbot:

Diffbot is a robot that examines the Web using computer vision and natural language processing, and provides developers with robust tools to find, extract and understand the objects from any Web page for use in their applications. Thousands of developers and businesses rely on Diffbot APIs to create consumer-friendly applications that use visual interpretation of the Web to re-imagine search, the mobile web and hundreds of other consumer applications. Customers include Adobe, CBS Interactive, DuckDuckGo, Cisco, eBay, Instapaper, Salesforce, Samsung, StumbleUpon. It is based in Palo Alto, CA.