Diana Pérez, Universidad Autonoma de Madrid; Alfio Gliozzo and Carlo Strapparava, Istituto per la Ricerca Scientifica e Tecnologica (IRST); Enrique Alfonseca and Pilar Rodriguez, Universidad Autonoma de Madrid; and Bernardo Magnini, Istituto per la Ricerca Scientifica e Tecnologica (IRST)
In previous work we have proved that the BLEU algorithm, originally devised for evaluating Machine Translation systems, can be applied to assessing short essays written by students. In this paper we present a comparative evaluation between this BLEU-inspired algorithm and a system based on Latent Semantic Analysis. In addition we propose an effective combination schema for them. Despite the simplicity of these shallow NLP methods, they achieve state-of-the-art correlations to the teachers’ scores while keeping the language-independence and without requiring any domain specific knowledge.