Automatic Identification of Aspectual Classes across Verbal Readings

Proceedings of the Fifth Joint Conference on Lexical and Computational Semantics (*SEM 2016), pages 12–22, Berlin, Germany, August 11-12, 2016. Automatic Identification of Aspectual Classes across Verbal Readings Ingrid Falk and Fabienne Martin University of Stuttgart firstname.lastname@ling.uni-stuttgart.de Abstract The automatic prediction of aspectual classes is very challenging for verbs whose aspectual value varies across readings, which are the rule rather than the excep- tion. This paper sheds a new perspective on this problem by using a machine learn- ing approach and a rich morpho-syntactic and semantic valency lexicon. In con- trast to previous work, where the aspec- tual value of corpus clauses is determined on the basis of features retrieved from the corpus, we use features extracted from the lexicon, and aim to predict the aspectual value of verbal readings rather than verbs. Studying the performance of the classifiers on a set of manually annotated verbal read- ings, we found that our lexicon provided enough information to reliably predict the aspectual value of verbs across their read- ings. We additionally tested our predic- tions for unseen predicates through a task based evaluation, by using them in the automatic detection of temporal relation types in TempEval 2007 tasks for French. These experiments also confirmed the re- liability of our aspectual predictions, even for unseen verbs. 1 Introduction It is well known that the aspectual value of a sen- tence plays an important role in various NLP tasks, like for instance the assessment of event factual- ity (Saurí and Pustejovsky, 2012), automatic sum- marisation (Kazantseva and Szpakowicz, 2010), the detection of temporal relations (Costa and Branco, 2012) or machine translation (Meyer et al., 2013). Since, however, the aspectual value of a sentence results from a complex interplay between lexical features of the predicate and its linguistic context, the automatic detection of this aspectual value is quite challenging. Studies on the computational modelling of as- pectual classes emerged about two decades ago with the work of Passonneau (1988) and Klavans and Chodorow (1992), among others. In proba- bly the most extensive study on the field, Siegel and McKeown (2000) extract clauses from a cor- pus and classify them into states and events, sort- ing the latter into culminated and non-culminated events in a subsequent step. The classification is based on features inspired by classic Vendlerian aspectual diagnostics, themselves collected from the corpus. Since, however, these features are collected on a type level, this method does not give satisfying results for verbs whose aspectual value varies across readings (henceforth ‘aspectu- ally polysemous verbs’), which are far from ex- ceptional (see section 3)1. This problem is directly addressed by Zarcone and Lenci (2008). These authors classify cor- pus clauses into the four Vendlerian aspectual cat- egories (states, activities, accomplishments and achievements), and like Siegel and McKeown, base their classification on (classic aspectual) fea- tures collected from the corpus. However, they ad- ditionally employ some syntactic properties of the predicate, a move that enables them to better ac- count for the influence of the linguistic context …

Content truncated to 500 words