Pretrained static word embeddings for Portuguese (BR+PT), trained by NILC on a large multi-genre corpus (~1.39B tokens, 17 sources).
AI & ML interests
Research topics of interest include sentiment analysis and opinion mining, text summarization, text simplification, reading and writing supporting tools, machine translation, corpus linguistics, computational terminology, parsing methods for varied linguistic levels (part of speech, syntax, semantics and discourse), automated evaluation, speech processing, and application of machine learning techniques and varied modeling strategies for language processing solutions, among several others.