PRISMATIC: Inducing Knowledge from a Large Scale Lexicalized Relation Resource

One of the main bottlenecks in natural language processing is the lack of a comprehensive lexicalized relation resource that contains fine grained knowledge on predicates. In this paper, we present PRISMATIC, a large scale lexicalized relation resource that is automatically created over 30 gb of text. Specifically, we describe what kind of information is collected in PRISMATIC and how it compares with existing lexical resources. Our main focus has been on building the infrastructure and gathering the data. Although we are still in the early stages of applying PRISMATIC to a wide variety of applications, we believe the resource will be of tremendous value for AI researchers, and we discuss some of potential applications in this paper.