Vietnamese Students’ Feedback Corpus

Vietnamese Students’ Feedback Corpus (UIT-VSFC) is the resource consists of over 16,000 sentences which are human-annotated with two different tasks: sentiment-based and topic-based classifications. To assess the quality of our corpus, we measure the annotator agreements and classification evaluation on the UIT-VSFC corpus.

As a result, we obtained the inter-annotator agreement of sentiments and topics with more than over 91% and 71% respectively. In addition, we built the baseline model with the Maximum Entropy classifier and achieved approximately 88% of the sentiment F1-score and over 84% of the topic F1-score.