研究人员开发了一个系统，用于自动对匈牙利学生论文中的反思水平进行分类，填补了该语言在自动化分析方面的空白。该研究使用了包含 1,954 篇论文的数据集，并尝试了传统的机器学习模型和匈牙利语特有的 transformer 模型。虽然经典方法取得了 71% 的较高总体得分，但 transformer 模型在少数类别上表现出更好的泛化能力，得分为 68%。研究结果表明，对于资源匮乏的语言，经典方法仍然具有相关性，而 transformer 模型则为不平衡数据集提供了鲁棒性。 AI
影响 为形态丰富、资源匮乏的语言的自动化文本分析提供了新的数据集和见解。
排序理由 学术论文，提出了一种新的数据集和方法论，用于自动化文本分类。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2605.02402v1 Announce Type: new Abstract: Reflective thinking is a key competency in education, but assessing reflective writing remains a time-consuming and subjective task for education experts. While automated reflective analysis has been explored in several languages, H…
Reflective thinking is a key competency in education, but assessing reflective writing remains a time-consuming and subjective task for education experts. While automated reflective analysis has been explored in several languages, Hungarian language was not researched extensively…
Reflective thinking is a key competency in education, but assessing reflective writing remains a time-consuming and subjective task for education experts. While automated reflective analysis has been explored in several languages, Hungarian language was not researched extensively…