Multimodal Building of Monolingual Dictionaries for Machine Translation by Non-Expert Users
Miquel Esplà-Gomis, Víctor M. Sánchez-Cartagena, Juan Antonio Pérez-Ortiz · 2011
This paper explores a new approach to help non-expert users with no background in linguistics to add new words to a monolingual dictionary in a rule-based machine translation system. Our method aims at obtaining the correct paradigm which explains not only the particular surface form introduced by the user, but also the rest of inflected forms of the word. An initial set of potential paradigms is automatically obtained and then interactively refined by the user with a novel graphical interface through active machine learning. We show the promising results of experiments performed with a Spanish monolingual dictionary. 1