oa Identificação automática de elementos básicos da frase latina: o Projecto OLISSIPO

The development of the information society imposes a more and more analytical computational management of the data. In the linguistic field, lemmatization is the preliminary operation for this purpose. The Latin lemmatizer LEMLAT is able to perform such operation placing a form received in input alongside its corresponding lemma. The current needs of the web and, more generally speaking, of the recent technologies of information retrieval (e-learning, philological and linguistic search, summarization) require the development of the LEMLAT capacities. In this paper, the authors present a proposal for the development of the system, in view of a complete morphological analysis of the word forms of a Latin text.