Feature Extraction and Classification

From String

CLAVIS is an automatic feature extraction and text classification system for European Portuguese. It seeks to assist the selection of adequate reading materials for teaching European Portuguese, especially as a second language. The system adopts the official classification presented by Framework for Teaching Portuguese Abroad, QUAREPE, with five-levels scale (A1, A2, B1, B2, and C1), while still allowing the use of a simplified 3-level scale (A, B, C).
CLAVIS uses lexical and syntactic information produced by the natural language processing chain STRING, as well as the hyphenator YAH developed by L2F-INESC-ID Lisboa. The CLAVIS was developed as part of a dissertation in Information Systems and Computer Engineering by Pedro Curto (2014). To quote, use the reference below:
Curto, P.: Classificador de textos para o ensino de português como segunda língua. Master’s thesis, Instituto Superior Técnico - Universidade de Lisboa, Lisboa (2014) bibtex.