Statistical language understanding using frame semantics
Daniel Gildea · 2001
Statistical Language Understanding Using Frame Semantics by Daniel Joseph Gildea Doctor of Philosophy in Computer Science University of California, Berkeley Professor Nelson Morgan, Chair We presentasystem for identifying the semantic relationships, or semantic roles, #lled by constituents of a sentence within a semantic frame. We use frame semantics as a level of representation intermediate between task-speci#c templates commonly used in information extraction and complete theories of language understanding using complex semantic structures. The system is based on statistical classi#ers trained on roughly 50,000 sentences that were hand-annotated with semantic roles bytheFrameNet semantic labeling project. We then parsed each training sentence into a syntactic tree and extracted various lexical and syntactic features, including the phrase typeofeach constituent, its grammatical function, and position in the sentence. These features were combined with knowledge of the predicate verb, noun, or adjective, as well as information such as the prior probabilities of various combinations of semantic roles. We used various lexical clustering algorithms to generalize across possible #llers of roles. Test sentences were parsed, were annotated with these features, and were then passed through the classi#ers. Our study also allowed us to compare the usefulness of di#erent features and feature-combination methods in the semantic role labeling task. We also explore the integration of role labeling with statistical syntactic parsing, and attempt to generalize among data for di#erent predicates. i Contents 1