Lessons Learned Applying NLP to Create a Web-Scale Knowledge Graph
High-quality Knowledge Graphs still mostly rely on structured data curated by humans. Such reliance on human curation is a major obstacle to the creation of a comprehensive, always-up-to-date knowledge graph of human knowledge.
This talk will discuss the lessons learned applying NLP to create the Diffbot Knowledge Graph, the largest autonomous knowledge graph of the public web. Diffbot crawls, extracts, and integrates information from approximately 80 billion web pages automatically, serving over 400 enterprise customers.
Vice President of Research at Diffbot
Filipe Mesquita is the Vice President of Research at Diffbot. He is interested in advancing the state-of-the-art in automated knowledge graph construction, particularly from unstructured documents.
He received his Ph.D. from the University of Alberta on the topic of automatically discovering and extracting relationships among named entities from text.
His work has been published in top-tier conferences and journals, such as EMNLP, NAACL, SIGMOD, TWEB, and SIGIR.