计算机与现代化 ›› 2024, Vol. 0 ›› Issue (06): 8-13.doi: 10.3969/j.issn.1006-2475.2024.06.002

• 人工智能 • 上一篇    下一篇

基于语义特征融合的作文自动评分方法

  



  1. (新疆师范大学计算机科学技术学院,新疆 乌鲁木齐 830054)
  • 出版日期:2024-06-30 发布日期:2024-07-17
  • 作者简介:袁航(1999—),男,河南周口人,硕士研究生,研究方向:自然语言处理,作文评分,E-mail:y503744685@163.com;杨勇(1979—),男,陕西汉中人,教授,博士,研究方向:自然语言处理、软件工程,E-mail:68523593@qq.com; 任鸽(1986—),女,河南兰考人,副教授,研究方向:数据挖掘,网络信息安全,E-mail:236789497@qq.com; 通信作者:帕力旦·吐尔逊(1970—),女(维族),新疆乌鲁木齐人,副教授,博士,研究方向:多模态信息处理,E-mail:pldtrs@xjnu.edu.cn。
  • 基金资助:
    新疆维吾尔自治区自然科学基金项目(2021D01B72); 国家自然科学基金资助项目(62167008,62066044)

Automatic Scoring Method for Composition Based on Semantic Feature Fusion



  1. (School of Computer Science and Technology, XinJiang Normal University, Urumqi 830054, China)
  • Online:2024-06-30 Published:2024-07-17

摘要: 摘要:作文自动评分技术是一种利用机器学习进行自然语言处理的技术。目前,基于深度学习的端到端模型在作文自动评分领域已经广泛使用。然而,由于端到端模型难以获取不同特征之间的相关性,因此提出一种基于语义特征融合的作文自动评分方法(TSEF)。该方法主要分为特征提取和特征融合2个阶段。特征提取阶段,使用Bert模型对输入文本进行预训练,并使用多头注意力机制对输入文本进行自训练,以补充预训练的不足;特征融合阶段,使用交叉融合方法将获取的不同特征融合,以此获得更好性能的模型。在实验中,将TSEF与许多强基线进行比较,结果表明了本文方法的有效性和稳健性。


关键词: 关键词:作文自动评分, 自训练, 预训练, 交叉融合

Abstract:
Abstract: Automatic composition scoring technology is a kind of natural language processing technology using machine learning. At present, end-to-end models based on deep learning have been widely used in the field of automatic essay scoring. However, because of the difficulty in obtaining correlations between different features in end-to-end models, Automatic Scoring Method for Composition Based on Semantic Feature Fusion (TSEF) has been proposed. This method is mainly divided into two stages: feature extraction and feature fusion. In the feature extraction stage, the Bert model is used to pre-train the input text, and a multi-head-attention mechanism is used to self-train the input text to supplement the shortcomings of pre-training; In the feature fusion stage, cross fusion methods are used to fuse the different features obtained in order to obtain a better performance model. In the experiment, TSEF was compared with many strong baselines, and the results demonstrated the effectiveness and robustness of our method.

Key words: Key words: automatic grading of essays, self-training, pre-training, cross fusion

中图分类号: