IJCAI 2013 Tutorial

Endowing Machines with WebSense

Loizos Michael
Computational Cognition Lab
Open University of Cyprus
loizos@ouc.ac.cy

 

For the first time in human history, the Web makes it possible for so much of the human knowledge to be available to so many. At the same time, natural language text, the very representation that facilitates this proliferation of knowledge sharing, acts as the main stumbling block for machines to make sense of this knowledge in a massive and unsupervised manner. Natural Language Processing tools and techniques are arguably central in a system trying to make sense of text available on the Web. Such tools aid in identifying the syntactic and semantic structure of individual sentences or groups of sentences, and represent it in a form appropriate both for human consumption, but equally importantly, for machine consumption. But then what?

Even if one assumes the use of state-of-the-art NLP parsers for this task, one may do much more in terms of exploiting knowledge available on the Web. In one direction, one may use the massive amounts of text to improve the quality of the parsed information. In a second direction, one may seek regularities in the information to identify entities of interest and facts about these entities. In a third direction, one may attempt to extract the underlying knowledge (which one could dub “websense” in reference and contrast to “commonsense”) that governs the available information, so that one may draw inferences about entities and facts that are only implicitly encoded on the Web. This tutorial will review some of the ongoing work in these directions.


Presenter's Short Bio: Loizos Michael is an assistant professor at Open University of Cyprus, where he founded and directs the Computational Cognition Lab. He was educated at University of Cyprus, where he received a B.Sc. in Computer Science with a minor degree in Mathematics. He continued his education at Harvard University, where he received an M.Sc. and a Ph.D. in Computer Science. His research seeks to understand cognitive processes often associated with individual or collective intelligence — such as learning, reasoning, sensing, communication, cooperation — and how those are employed by humans and other organisms in everyday life. Emphasis in this research is placed on the development of formal computational models for various aspects of cognitive processes, and the analysis of the formal implications that such models have. The computational view and treatment of cognition is complemented by simulations, real-world experiments, and psychological studies, designed to validate the proposed models and to identify features thereof that warrant further investigation.