We investigate in this paper the degree of overlap between synonym sets of translated word pairs across three languages: French, English and Romanian. We use for this purpose a French Synonym Dictionary, a Romanian Synonym Dictionary, Princeton’s WordNet and Google Translate API. We build a database containing pairs of (translated) words from the three languages, along with their corresponding ...

This paper discusses particular linguistic challenges in the task of compiling the Brazilian Portuguese Wordnet, the Wordnet.Br. After setting the scene by overviewing methodological issues, it focuses on the basic steps taken to compile the Wordnet.Br core database: a machine-tractable thesaurus-like lexical database. The discussion is split between three domains: the Linguistic Domain, the Re...

This paper describes a framework for recognizing contradictions between multiple text sources by relying on three forms of linguistic information: (a) negation; (b) antonymy; and (c) semantic and pragmatic information associated with the discourse relations. Two views of contradictions are considered, in which a novel method of recognizing contrast and of finding antonymies are described. Contr...

This paper describes a method of alignment between the noun part of WordNet and Goi-Taikei’s Ontology. The result of the automatic alignment shows that matching words in Japanese as well as in English is highly efficient in extracting correct matches. Analyses of correct alignments by hand illustrate that, in ontology alignment, it is not always possible to preserve relative relations between c...

In this paper a new approach for achieving the conceptual matching between user queries and web documents is presented. The key of the proposed system is to use FIS-CRM (Fuzzy Interrelations and Synonymy Concept Representation Model) to represent the indexed web pages. This model (also implemented in the FISS metasearcher) is supported by a fuzzy synonymy dictionary and various thematic fuzzy o...