A Joint Sequential and Relational Model for Frame-Semantic Parsing

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 1247–1256 Copenhagen, Denmark, September 7–11, 2017. c⃝2017 Association for Computational Linguistics A Joint Sequential and Relational Model for Frame-Semantic Parsing Bishan Yang Machine Learning Department Carnegie Mellon University bishan@cs.cmu.edu Tom Mitchell Machine Learning Department Carnegie Mellon University tom.mitchell@cs.cmu.edu Abstract We introduce a new method for frame- semantic parsing that significantly im- proves the prior state of the art. Our model leverages the advantages of a deep bidi- rectional LSTM network which predicts semantic role labels word by word and a relational network which predicts seman- tic roles for individual text expressions in relation to a predicate. The two net- works are integrated into a single model via knowledge distillation, and a unified graphical model is employed to jointly decode frames and semantic roles during inference. Experiments on the standard FrameNet data show that our model sig- nificantly outperforms existing neural and non-neural approaches, achieving a 5.7 F1 gain over the current state of the art, for full frame structure extraction. 1 Introduction One way to represent meaning is through organi- zation of semantic structures. Consider the follow- ing sentences “John sells Marry a car.” and “Mary buys a car from John.”. While having different syntactic structures, they express the same type of event that involves a buyer, a seller, and goods. Such meaning can be represented using seman- tic frames – structured representations that char- acterize events, scenarios, and the participants. Researchers have developed FrameNet (Baker et al., 1998; Fillmore et al., 2003), a large lexi- cal database of English that comes with sentences annotated with semantic frames. It has been con- sidered a valuable resource for Natural Language Processing and useful for studying tasks such as information extraction, machine translation, and question answering (Surdeanu et al., 2003; Shen and Lapata, 2007; Liu and Gildea, 2010). Here we consider the task of automatic extrac- tion of semantic frames as defined in FrameNet. This include target identification – identifying frame-evoking predicates, frame identification – identifying which frame each predicate evokes, and semantic role labeling (SRL) – identifying phrasal arguments of each evoked frame and la- beling them with the frame’s semantic roles. Con- sider the sentence “We decided to treat the pa- tient with combination chemotherapy.”. Here “de- cided” evokes the DECIDING frame and “treat” evokes the CURE frame. Each frame takes a set of arguments that fill the semantic roles of the frame, as illustrated below: [WeCOGNIZER] decided [to treat the patient with combination chemotherapyDECISION]. [WeHEALER] decided to treat [the patientPATIENT] [with combination chemotherapyTREATMENT]. We address frame identification and semantic role labeling in this work.1 Frame identification can be addressed as a word sense disambiguation problem, while semantic role labeling can be for- mulated as a structured prediction problem. We train different neural network models for these two problems, and interpret their outputs as factors in a graphical model for performing joint inference over the distribution of frames and semantic roles. Specifically, our frame identification model is …

Content truncated to 500 words