Semantic Textual Similarity Evaluation for Spanish Sentences

This project sought to enhance the natural language processing research of the MTA SZTAKI institute in Budapest, Hungary, by extending their semantic textual similarity system to evaluate Spanish sentences. Language analysis resources were collected to generate a working system to analyze similarities between Spanish sentence pairs. This system was based on that which the institute had previously developed for English. The final system was tested against large data sets of sentence pairs, and compared to a Gold Standard of scores created by human linguists, with the goal of having a high correlation between the two data sets