This is the authors' abstract. We don't add key points for this paper.
The attributes, such as knowledge points and difficulty, are important parts of questions. It is related to the automatic classification, auto-generating test paper and adaptive learning. The traditional labeling, however, is done by domain experts and teachers with low efficiency which make it difficult to achieve large-scale automatic labeling. Automatic labeling not only saves a lot of manual labeling costs, but also executes question labeling in batches. Question labeling is an essential text multi-label classification problem. In this paper, we propose an improved multi-label classification algorithm based on naive Bayes algorithm, combined with AdaBoost algorithm. We develop a new evaluation model of classification precision for the hierarchical characteristic of question knowledge point structure. The experimental results demonstrate the proposed algorithm outperforms to the naive Bayes algorithm and the SVM algorithm in precision, recall and Fl- value.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Sun et al. (2018) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: