RTV: Tree Kernels for Thematic Role Classification

Proceedings of the 4th International Workshop on Semantic Evaluations (SemEval-2007), pages 288–291, Prague, June 2007. c⃝2007 Association for Computational Linguistics RTV: Tree Kernels for Thematic Role Classification Daniele Pighin FBK-irst; University of Trento, DIT pighin@itc.it Alessandro Moschitti University of Trento, DIT moschitti@dit.unitn.it Roberto Basili University of Rome Tor Vergata, DISP basili@info.uniroma2.it Abstract We present a simple, two-steps supervised strategy for the identification and classifica- tion of thematic roles in natural language texts. We employ no external source of in- formation but automatic parse trees of the in- put sentences. We use a few attribute-value features and tree kernel functions applied to specialized structured features. The result- ing system has an F1 of 75.44 on the Se- mEval2007 closed task on semantic role la- beling. 1 Introduction In this paper we present a system for the labeling of semantic roles that produces VerbNet (Kipper et al., 2000) like annotations of free text sentences us- ing only full syntactic parses of the input sentences. The labeling process is modeled as a cascade of two distinct classification steps: (1) boundary detection (BD), in which the word sequences that encode a thematic role for a given predicate are recognized, and (2) role classification (RC), in which the type of thematic role with respect to the predicate is as- signed. After role classification, a set of simple heuristics are applied in order to ensure that only well formed annotations are output. We designed our system on a per-predicate basis, training one boundary classifier and a battery of role classifiers for each predicate word. We clustered all the senses of the same verb together and ended up with 50 distinct boundary classifiers (one for each target predicate word) and 619 role classifiers to rec- ognize the 47 distinct role labels that appear in the training set. The remainder of this paper is structured as fol- lows: Section 2 describes in some detail the archi- tecture of our labeling system; Section 3 describes the features that we use to represent the classifier examples; Section 4 describes the experimental set- ting and reports the accuracy of the system on the SemEval2007 semantic role labeling closed task; fi- nally, Section 5 discusses the results and presents our conclusions. 2 System Description Given a target predicate word in a natural language sentence, a SRL system is meant to correctly iden- tify all the arguments of the predicate. This problem is usually divided in two sub-tasks: (a) the detection of the boundaries (i. e. the word span) of each argu- ment and (b) the classification of the argument type, e.g. Arg0 or ArgM in PropBank or Agent and Goal in FrameNet or VerbNet. The standard approach to learn both the detection and the classification of predicate arguments is sum- marized by the following steps: 1 Given a sentence from the training-set, gener- ate a full syntactic parse-tree; 2 let P and A be the set of predicates and the set of parse-tree nodes (i.e. the potential argu- ments), respectively; 3 for …

Content truncated to 500 words