Gleaning Information from the Web: Using Syntax to Filter Out Irrelevant Information

摘要: In this paper, we describe a system called Glean, which is predicated on the idea that any coherent text contains significant latent information, such as syntactic structure and patterns of language use, can be used to enhance perlbrmauce Information Retrieval systems. We propose an approach information retrieval makes use obtained using tool supertagger. A supertagger corpus training material semi-automatically induce call augmented-patterns. show how these augmented may along with standard Web search engine or IR retrieve identify relevant filter out irrelevant items. experiment in domain official appointments, where are shown reduce number potentially documents by upwards 80%.