We present an unsupervised learning method for disambiguating verbs that belong to more than one Levin verb class (1993) when occurring in a particular syntactic frame. We used examples that contain unambiguous verbs in each verb class as the training data for ambiguous verbs in that class. A Naive Bayesian classifier was employed for the disambiguation task using context words as features. Our...

When presented with a novel verb in a transitive frame (X is Ving Y), young children typically select a causative event referent, rather than one in which agents engage in parallel, non-causative synchronous events. However, when presented with a conjoined-subject intransitive frame (X and Y are Ving), participants (even adults, as we show) are at chance. Although in some instances, children ol...

In a majority of languages, the time of an event is expressed by marking tense on the verb. There is substantial evidence that the production of verb tense in sentences is more severely impaired than other functional categories in persons with agrammatic aphasia. The underlying source of this verb tense impairment is less clear, particularly in terms of the relative contribution of conceptual-s...

This paper presents a semi-automatic method for identifying terms that evoke semantic frames (Fillmore, 1982). The method is tested as a means of identifying lexical units that can be added to existing frames or to new, related frames, using a large corpus on the environment. It is hypothesized that a method based on distributional semantics, which exploits the assumption that words that appear...

We develop a probabilistic latent-variable model to discover semantic frames—types of events or relations and their participants—from corpora. Our key contribution is a model in which (1) frames are latent categories that explain the linking of verb-subject-object triples in a given document context; and (2) cross-cutting semantic word classes are learned, shared across frames. We also introduc...

Decades of research have documented that acquiring the meanings of verbs is, on average, more difficult than acquiring the meanings of nouns. This is likely because verb learning requires more information, and information of a different kind, than noun learning. One particularly powerful source of information for verb learning is linguistic context. For example, Naigles (1990) demonstrated in a...

Children have difficulty comprehending novel verbs in the double object dative (e.g., Fred blicked the dog a stick) as compared to the prepositional dative (e.g., Fred blicked a stick to the dog). We explored this pattern with 3 and 4 year olds (N = 60). In Experiment 1, we replicated the documented difficulty with the double object frame, even though we provided more contextual support. In Exp...

There are a number of well-known restrictions for the Dative Alternation (cf. Green (1974), Oehrle (1976), Gropen, Pinker, Hollander, & Goldberg (1989), Pinker (1989), Pesetsky (1992), Levin (1993). I will show that several of the low-level semantic restrictions are consequences of a more general one involving the incorporation of a manner component into the meaning of the verb. These restricti...