试题知识点分类是智慧教育中的一个核心技术支撑。传统试题知识点分类方法往往忽略了试题图片与试题文本之间的深层语义关联。针对上述问题,该文提出了一种基于多模态学习的试题知识点分类方法。该方法鉴于不同模态的试题特征之间存在互补关系,采用协同注意力机制(Co-attention)分别获取试题文本引导的试题图片特征和试题图片引导的试题文本特征;然后通过门控机制动态地对两者的特征进行融合表示,以获取更丰富的试题语义信息。实验结果表明,在某教育机构提供的物理试题数据集上,一级知识点和二级知识点的分类准确率可以分别达到95.09%和83.18%,Macro-F1值可以分别达到64.20%和50.63%。通过分析发现,多模态学习能有效弥补小样本试题知识点分类中的特征稀疏问题。因此,该方法可有效支撑智慧教育中的试题知识点分类。
Abstract
The classification of knowledge points in test questions is a core technical support in smart education. To capture the deep semantic relationship between test pictures and test questions, this article proposes a multimodal learning based classification method for knowledge points in test text. To utilize the complementary relationship between different modalities, the cooperative attention mechanism (Co-attention) is used to obtain the text-guided picture features of test questions and the picture-guided text features of test questions, respectively. Then, the two features are dynamically fused through the gating mechanism to obtain richer semantic information of the test questions. Experiments on the physics test dataset show that the classification accuracy of the first-level knowledge points and the second-level knowledge points can reach 95.09% and 83.18%, respectively; the Macro-F1 value can reach 64.20% and 50.63%, respectively.
关键词
知识点分类 /
文本卷积神经网络 /
多模态融合 /
协同注意力机制
{{custom_keyword}} /
Key words
knowledge points classification /
TextCNN /
multimodal fusion /
co-attention mecharism
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 教育部关于印发《教育信息化2.0行动计划》的通知[J].中华人民共和国教育部公报,2018(04): 118-125.
[2] 《2020年教育信息化和网络安全工作要点》发布[J].中国教育网络,2020(04): 12-13.
[3] 汪忠国,吴敏.基于向量空间模型的题库相似度检查算法[J].计算机系统应用,2010,19(03): 213-216.
[4] 黄仔. 基于多模态学习的试题建模方法与应用研究[D].合肥: 中国科学技术大学硕士学位论文, 2019.
[5] 葛华丽.基于知识点的试题库组卷算法[J].电脑知识与技术,2009,5(26): 7497-7498.
[6] 杨志明.题库建设中的编题与组卷功能[J].教育测量与评价(理论版),2016(02): 4-7.
[7] 曾沛,周霞.高校试卷评估评价系统设计研究[J].电脑知识与技术,2017,13(30): 96-97.
[8] 朱天宇,黄振亚,陈恩红,等.基于认知诊断的个性化试题推荐方法[J].计算机学报,2017,40(01): 176-191.
[9] SUNIL D, DOJA M N. Recommender system for personalized adaptive e-learning platforms to enhance learning capabilities of learners based on their learning style and knowledge level [C]//Proceedings of the International Conference on Sustainable Computing in Science, Technology & Management, 2019: 26-28.
[10] CICCARESE P, OCANA M, CLARK T. Open semantic annotation of scientific publications using DOMEO [C]//Journal of Biomedical Semantics, 2012, 3(1): 1-14.
[11] SALTON G, LESK M E. Computer evaluation of indexing and text processing [J]. Journal of the ACM, 1968, 15(1): 8-36.
[12] PLATT J. Sequential minimal optimization: A fast algorithm for training support vector machines [J]. Technical Report MSR-TR-98-14, Microsoft Research, 1998.
[13] 植兆衍,彭宏.基于向量空间模型的试题分类系统[J].计算机工程与设计,2008(12): 3227-3229.
[14] 朱刘影,杨思春.基于SVM的地理试题自动分类[J].计算机应用研究,2018,35(09): 2707-2710.
[15] 郭崇慧,吕征达.一种基于集成学习的试题多知识点标注方法[J].运筹与管理,2020,29(02): 129-136.
[16] 梁圣. 基于RNN的试题相似性检测与分类研究[D]. 株洲: 湖南工业大学硕士学位论文, 2018.
[17] 胡国平,张丹,苏喻,等.试题知识点预测: 一种教研知识强化的卷积神经网络模型[J].中文信息学报,2018,32(05): 137-146.
[18] 任泽裕,王振超,柯尊旺,等.多模态数据融合综述[J].计算机工程与应用,2021,57(18): 49-64.
[19] LIU Q, HUANG Z, HUANG Z, et al. Finding similar exercises in online education systems [C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2018: 1821-1830.
[20] YIN Y, LIU Q, HUANG Z, et al. Quesnet: A unified representation for heterogeneous test questions [C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2019: 1328-1336.
[21] TRUONG Q T, LAUW H W. Vistanet: Visual aspect attention network for multimodal sentiment analysis [C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33(01): 305-312.
[22] HUANG F, ZHANG X, ZHAO Z, et al. Image-text sentiment analysis via deep multimodal attentive fusion [J]. Knowledge based Systems, 2019, 167: 26-37.
[23] WANG Y, QIAN S, HU J, et al. Fake news detection via knowledge-driven multimodal graph convolutional networks [C]//Proceedings of the International Conference on Multimedia Retrieval, 2020: 540-547.
[24] ZHANG Q, FU J, LIU X, et al. Adaptive co-attention network for named entity recognition in tweets [C]//Proceedings of the 32nd AAAI Conference on Artificial Intelligence, 2018: 5674-5681.
[25] DEVLEN J, CHANG M W, LEE K, et al. BERT: Pre training of deep bidirectional transformers for language understanding [C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2019: 4171-4186.
[26] KIM Y. Convolutional neural networks for sentence classification [C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2014: 1746-1751.
[27] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition [C]//Proceedings of the 3rd International Conference on Learning Representations, Computational and Biological Learning Society, 2015.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(62166007)
{{custom_fund}}
PDF(11244 KB)
