Probase : a Universal Knowledge Base for Semantic Search

Abstract

We demonstrate a prototype system that showcases the power of using a knowledge base (Probase) for search. The goal of Probase is to enable common sense computing, and its foundation is a universal, probabilistic ontology that is more comprehensive than any of the existing ontologies. Currently, it contains over 2.7 million concepts harnessed automatically from a corpus of 1.68 billion web pages. Unlike traditional knowledge bases that treat knowledge as black and white, it supports probabilistic interpretations of the information it contains. The probabilistic nature also enables it to incorporate heterogeneous information in a natural way. Besides the system, we also demonstrate two applications, i) semantic web search and ii) understanding and searching web tables, that are built on top of the Probase framework. They indicate that a little common sense goes a longway: machines can be made more intelligent if it has access to the right knowledge.