Automatic Labeling of Semantic Roles

Abstract

We present a system for identifying the semantic relationships, or semantic roles, filled by constituents of a sentence within a semantic frame. Given an input sentence and a target word and frame, the system labels constituents with either abstract semantic roles, such as Agent or Patient, or more domain-specific semantic roles, such as Speaker, Message, and Topic. The system is based on statistical classifiers trained on roughly 50,000 sentences that were hand-annotated with semantic roles by the FrameNet semantic labeling project. We then parsed each training sentence into a syntactic tree and extracted various lexical and syntactic features, including the phrase type of each constituent, its grammatical function, and its 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 fillers of roles. Test sentences were parsed, were annotated with these features, and were then passed through the classifiers. Our system achieves 82% accuracy in identifying the semantic role of presegmented constituents. At the more difficult task of simultaneously segmenting constituents and identifying their semantic role, the system achieved 65% precision and 61% recall. Our study also allowed us to compare the usefulness of different 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 to predicates unseen in the training data.

Efficient Convolution Kernels for Dependency and Constituent Syntactic Trees

“…These are schematic representations of situations involving various participants, properties and roles in which a word may The semantic roles Suspect and Authorities are specific to this Frame. The common approach to learn the classification of predicate arguments relates to the extraction of features from syntactic parse trees of the training sentences [15]. An alternative representation based on tree kernels selects the minimal partial tree that includes a predicate with only one of its arguments [6].…”

Bi-directional Joint Inference for Entity Resolution and Segmentation Using Imperatively-Defined Factor Graphs

“…Another common alternative is feedforward N-best lists, e.g. [20,21]. The feedforward probability distribution can be better approximated by sampling [2], but this flow of information is still uni-directional.…”

Towards Technology Structure Mining from Scientific Literature

“…Recognition [4], Semantic Role Identification [5], and Relation Extraction [6], [7]. Considering technology as applied science, then scientific publications can be considered as a primary source of information about technologies and emerging technological trends.…”