This paper describes our system for the classification of argument coercion for SemEval-2010 Task 7. We present two approaches to classifying an argument’s semantic class, which is then compared to the predicate’s expected semantic class to detect coercions. The first approach is based on learning the members of an arbitrary semantic class using WordNet’s hypernymy structure. The second approac...

This paper treats the classification of the semantic functions performed by adnominal constituents in Japanese, where many parts of speech act as adnominal constituents. In order to establish a formal treatment of the semantic roles, the similarities and differences among adnominal constituents, i.e. adjectives and "noun + NO (in English "of + noun")" structures, which have a broad range of sem...

One of the goals of this workshop was to begin to lay the foundations for a comprehensive framework for understanding and classifying different problems, approaches and techniques in the field of semantic integration and interoperability. Another way to look at this, is to create a map of the field. In this short paper, we give an example of what a region on such a map might look like. We consi...

We describe an approach to classifying word senses and semantic roles that focuses on the entailments that can be drawn from sentences. A key observation is that entailments arise from sentences, not word senses, and by taking a compositional approach to deriving entailments, we can significantly simplify the number of senses that are required. In addition, we argue that most uses of semantic r...

This paper presents a method for detecting compound nominalisations from open data, and providing a semantic intepretation. It uses a statistical model based on confidence intervals over frequencies extracted from a large, balanced corpus. Using three paraphrases of the given compound nominalisation, and interpretation preferences of its components, the algorithm achieves about 70% accuracy in ...

1. To stimulate collaboration between diverse communities bound by common objectives in the area of semantic interoperability and integration; 2. To lay the foundation for a framework and a theory for understanding and classifying technologies for semantic interoperability and integration; 3. To set the research agenda for this area with the long-term aim of building a “research pipeline” for c...

In this paper, a study on the classification performance of a vector space model (VSM) and of latent semantic indexing (LSI) applied to the task of spam filtering is summarized. Based on a feature set used in the extremely widespread, de-facto standard spam filtering system SpamAssassin, a vector space model and latent semantic indexing are applied for classifying e-mail messages as spam or not...

On the Semantic Desktop users maintain their model of the world in a formal personal information model ontology. Concepts from this ontology are used to annotate documents from desktop, allowing efficient navigation and browsing of these. However, the mental overhead required for correctly classifying new incoming document is substantial. We present the integration of the ontology-based informa...

There is a gap between low-level descriptions of image content and the semantic understanding of users to query image databases in the content-based image retrieval. In this paper, we put forward a method of classifying image regions hierarchically using their semantics and that resembles peoples’ perception more than using low-level features. The experiments show, the better precision of seman...