Professor Eduard Hovy received his Ph.D degree from Yale University. He is now a professor at the Language Technologies Institute of Carnegie Mellon University (CMU). He is Co-Director for Research of the Command, Control, and Interoperability Center for Advanced Data Analysis.
He is AAAI Fellow, ACL Fellow. He was president of several important international organizations including ACL, ATMA and DGSNA. He was editor-in-chief of Springer Verlag Book series on Theory and Applications of Natural Language Processing since 2010. He was general chair and program committee chair of large number of international conferences.
His research focuses on various topics, including aspects of the computational semantics of human language (such as text analysis, event detection and coreference, text summarization and generation, question answering, discourse processing, ontologies, text mining, text annotation, and machine translation evaluation), aspects of social media (such as event detection and tracking, sentiment and opinion analysis, and author profile creation), analysis of the semantics of non-textual information such as tables, and aspects of digital government.
内容摘要：Automated Machine Reading for self-learning has been a longstanding dream of AI. But no successful system NLP reading system has been built to date. A central problem is the fact that no text is ever complete in itself; every author relies on the reader’s background knowledge and inferential capabilities to fill in gaps and omissions and imprecisions in the writing. In this talk I describe three projects that focus on acquiring and using different kinds of background knowledge. The SAFT knowledgebase construction system is a state of the art information extraction engine. The PropStore is a structure built using distributional semantics vectors to represent the basic semantic meanings of words and word-senses. The Profile Engine is a system that uses information of 3 million people in Wikidata to create generic profiles that can function as expectations of people. These projects, developed at CMU and USC/ISI between 2010 and 2017, contribute to the kinds of knowledge that people bring to bear when they read and understand texts. Without this sort of capability and knowledge, automated semantic reading systems may never be able to become self-steered knowledge-gathering automata.
