ParkYoungWoun/vietnamese_students_feedback

越南学生反馈语料库（UIT-VSFC）是一个包含超过16,000个句子的资源，用于情感分析和主题分类的跨学科研究。每个句子都经过人工标注，情感分为负面、中性和正面三类，主题分为讲师、培训计划、设施和其他四类。数据集的标注者一致性在情感和主题上分别超过91%和71%。使用最大熵分类器作为基线模型，情感和主题分类的F1分数分别达到约88%和84%。数据集分为训练集（11426个样本）、验证集（1583个样本）和测试集（3166个样本）。

The Vietnamese Students’ Feedback Corpus (UIT-VSFC) is a resource consisting of over 16,000 sentences annotated for sentiment analysis and topic classification. Each sentence is human-annotated with sentiment (negative, neutral, positive) and topic (lecturer, training_program, facility, others) labels. The inter-annotator agreement for sentiment and topic annotations exceeds 91% and 71% respectively. A baseline model using the Maximum Entropy classifier achieved approximately 88% F1-score for sentiment classification and over 84% F1-score for topic classification. The dataset is split into training (11,426 samples), validation (1,583 samples), and test (3,166 samples) sets.