综合语义技术与LSTM神经网络的主观题自适应评卷方法

doi:10.16088/j.issn.1001-6600.2020060402

广西师范大学学报（自然科学版） ›› 2021, Vol. 39 ›› Issue (2): 51-61.doi: 10.16088/j.issn.1001-6600.2020060402

综合语义技术与LSTM神经网络的主观题自适应评卷方法

徐庆婷^1,2, 张兰芳^3*, 朱新华^1,2

1.广西师范大学计算机科学与信息工程学院, 广西桂林541004;
2.广西区域多源信息集成与智能处理协同创新中心, 广西桂林 541004;
3.广西师范大学教育学部, 广西桂林541004

收稿日期:2020-06-04 修回日期:2020-10-11 出版日期:2021-03-25 发布日期:2021-04-15
通讯作者: 张兰芳(1964—),女,广西柳州人,广西师范大学教授。E-mail:lfzhang64@163.com
基金资助:
国家自然科学基金(61967003,62062012);广西自然科学基金(2020GXNSFAA159082);广西多源信息挖掘与安全重点实验室系统性研究课题(20-A-03-02)

An Automatic Scoring Method for Subjective Questions Using Semantic Technologies and LSTM

XU Qingting^1,2, ZHANG Lanfang^3*, ZHU Xinhua^1,2

1. College of Computer Science and Information Engineering,Guangxi Normal University, Guilin Guangxi 541004, China;
2. Guangxi Collaborative Innovation Center of Multisource Information Integration and Intelligent Processing, GuilinGuangxi 541004, China;
3. Department of Education, Guangxi Normal University, Guilin Guangxi 541004, China

Received:2020-06-04 Revised:2020-10-11 Online:2021-03-25 Published:2021-04-15

摘要/Abstract

摘要： 针对已有主观题评卷存在的不足,本文提出了一种改进的综合语义技术与LSTM神经网络的主观题自适应评卷方法。首先,通过构建LSTM神经网络分类模型实现对问句试题语料集的题目类型和疑问类型同步双分类,为实现不同的问句类型采用不同的主观题自动评卷方法提供前期保障;然后,结合句法分析和依存关系分析对题目和答案文本进行剖析,包括通过对题目的依存关系分析得到疑问词在问句中的成分,再结合依存关系对答案进行剖析,得到标准答案和学生答案中各自的核心语义、非核心语义和否定语气词;最后,根据不同的问句类型,采用不同的主观题自动评卷方法,实现同一标准答案的不同学生回答方式的自适应评卷,提高主观题智能评卷的灵活性和准确性。

关键词: LSTM, 依存关系, 主观题, 自适应评卷

Abstract: Aiming at the shortcomings of the existing subjective question marking, this paper proposes an automatic scoring method for subjective questions using semantic technology and LSTM. Firstly, the LSTM neural network classification model is constructed to realize the simultaneous double classification of the title type and question type of the subjective question corpus. It provides the early guarantee for the realization of different question types using different automatic scoring methods; Then, syntax analysis and dependencies analysis are used to analyze questions and answers (standard answers and student’s answers), the components of interrogative words in questions are obtained through the dependencies analysis of title, and the core semantic, the non-core semantic and the negative tone of the answers are obtained through the dependencies analysis of the answers. Finally, for different types of questions, different automatic marking methods of subjective questions are adopted to realize the self-adaptive marking of different students with the same standard answers, which further improves the flexibility and accuracy of the subjective question scoring system.

Key words: LSTM, dependencies, subjective questions, self-adaptive scoring

中图分类号:

TP391

徐庆婷, 张兰芳, 朱新华. 综合语义技术与LSTM神经网络的主观题自适应评卷方法[J]. 广西师范大学学报（自然科学版）, 2021, 39(2): 51-61.

XU Qingting, ZHANG Lanfang, ZHU Xinhua. An Automatic Scoring Method for Subjective Questions Using Semantic Technologies and LSTM[J]. Journal of Guangxi Normal University(Natural Science Edition), 2021, 39(2): 51-61.

参考文献

[1] 刘伟,亓子森,王目宣.主观题自动测评研究[J].北京邮电大学学报(社会科学版),2016,18(4):108-116.
[2] 冯光,乔丹丹,常静怡.基于分词匹配的主观题自动评阅技术研究[J].计算机与现代化,2013,1(3):212-214,219.DOI:10.3969/j.issn.1006-2475.2013.03.054.
[3] 陈贤武,刘道波.基于语句相似度的主观试题自动阅卷模型研究[J].武汉大学学报(工学版),2018,51(7):654-658.DOI: 10.14188/j.1671-8844.2018-07-014.
[4] RUDNER L M,GAGNE P.An overview of three approaches to scoring written essays by computer[J].Practical Assessment,Research,and Evaluation,2000,7(1):26.DOI: 10.7275/m6xa-zg39.
[5] CURRAN M,DRAUS P,MARUSCHOCK G.Student perceptions of project essay grade (PEG) software[J].Issues in Information Systems,2013,14(1):89-98.
[6] LANDAUER T K,FOLTZ P W,LAHAM D.An introduction to latent semantic analysis[J].Discourse Processes,1998,25(2/3):259-284.DOI: 10.1080/01638539809545028.
[7] CALLEAR D H,JERRAMS-SMITH J,SOH V.CAA of short non-MCQ answers[C]//Proceedings of the 5th International Computer Assisted Assessment Conference.Loughborough:Loughborough University,2001:1-14.
[8] MITCHELL T,RUSSELL T,BROOMHEAD P,et al.Towards robust computerised marking of free-text responses[C]//Proceedings of the 6th International Computer Assisted Assessment Conference.Loughborough:Loughborough University,2002:233-249.
[9] BURSTEIN J,LEACOCK C,SWARTZ R.Automated evaluation of essays and short answers[C]//Proceedings of the 5th International Computer Assisted Assessment Conference.Loughborough:Loughborough University,2001:1-13.
[10] BURSTEIN J,TETREAULT J,MADNANI N.The e-rater automated essay scoring system[M]//SHERMIS M D,BURSTEIN J.Handbook of automated essay evaluation:Current applications and new direction.New York:Routledge,2013:55-67.
[11] 苏方方.基于领域本体和句框架的主观题自动阅卷算法研究[D].桂林:广西师范大学,2012.
[12] 张胜楠.基于文本相似度的短文本主观题自动评分方法研究[D].武汉:武汉理工大学,2017.
[13] 刘逸雪,卢雨轩,丁亮,等.基于Bi-LSTM的数学主观题自动阅卷方法[J].管理观察,2018(2):109-113.DOI: 10.3969/j.issn.1674-2877.2018.02.044.
[14] 张翠翠.基于双向遍历空间模型的主观题评判算法设计和阅卷系统实现[D].合肥:合肥工业大学,2018.
[15] 陈帅帅.基于文本相似度的主观题智能阅卷技术研究[D].北京:北京工业大学,2019.
[16] KOBAYASHI V B,MOL S T,BERKERS H A,et al.Text classification for organizational researchers::a tutorial[J].Organizational Research Methods,2018,21(3):766-799.DOI: 10.1177/1094428117719322.
[17] McCALLUM A,NIGAM K.A comparison of event models for naive Bayes text classification[C]//Proceedings of the AAAI-98 Workshop on Learning for Text Categorization.Palo Alto,CA:AAAI Press,1998:41-48.
[18] MANEK A S,SHENOY P D,MOHAN M C,et al.Aspect term extraction for sentiment analysis in large movie reviews using Gini Index feature selection method and SVM classifier[J].World Wide Web,2017,20(2):135-154.DOI: 10.1007/s11280-015-0381-x.
[19] 李敏,余正涛.结合加权特征向量空间模型和RBPNN的文本分类方法[J].计算机系统应用,2012,21(12):85-89,71.DOI: 10.3969/j.issn.1003-3254.2012.12.019.
[20] 周志华.机器学习[M].北京:清华大学出版社,2016:73-92.
[21] RUIZ M E,SRINIVASAN P.Hierarchical text categorization using neural networks[J].Information Retrieval,2002,5(1):87-118.DOI: 10.1023/A:1012782908347.
[22] HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.DOI: 10.1162/neco.1997.9.8.1735.
[23] GREFFK,SRIVASTAVA R K,KOUTNÍK J,et al.LSTM:a search space odyssey[J].IEEE Transactions on Neural Networks and Learning Systems,2017,28(10):2222-2232.DOI: 10.1109/TNNLS.2016.2582924.
[24] 熊涛.基于长短时记忆网络的多标签文本分类[D].杭州:浙江大学,2017.
[25] 胡新辰.基于LSTM的语义关系分类研究[D].哈尔滨:哈尔滨工业大学,2015.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

综合语义技术与LSTM神经网络的主观题自适应评卷方法

An Automatic Scoring Method for Subjective Questions Using Semantic Technologies and LSTM

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

Metrics

本文评价

推荐阅读 10

[1]	薛涛, 丘森辉, 陆豪, 秦兴盛. 基于经验模态分解和多分支LSTM网络汇率预测[J]. 广西师范大学学报（自然科学版）, 2021, 39(2): 41-50.
[2]	张明宇, 赵猛, 蔡夫鸿, 梁钰, 王鑫红. 基于深度学习的波浪能发电功率预测[J]. 广西师范大学学报（自然科学版）, 2020, 38(3): 25-32.
[3]	唐素勤, 黄运有, 王娜娜. 基于依存语法及本体技术的问句分析[J]. 广西师范大学学报（自然科学版）, 2014, 32(4): 52-58.