图书情报工作

图书情报工作 >

2019 , Vol. 63 >Issue 5: 92 - 99

DOI: https://doi.org/10.13266/j.issn.0252-3116.2019.05.011

情报研究

基于深度学习的MOOC论坛探索型对话识别方法研究

  • 董庆兴 ,
  • 李华阳 ,
  • 曹高辉 ,
  • 夏立新
展开
  • 1 华中师范大学信息管理学院 武汉 430079;
    2 武汉大学信息资源研究中心 武汉 430079;
    3 腾讯AI Lab 深圳 518057
董庆兴(ORCID:0000-0003-3512-9333),系主任,副教授,博士;李华阳(ORCID:0000-0003-3539-8648),助理研究员,本科;夏立新(ORCID:0000-0002-4162-2282),副校长,教授,博士

收稿日期: 2018-06-27

  修回日期: 2018-08-12

  网络出版日期: 2019-03-05

基金资助

本文系国家自然科学基金项目"面向群智感知大数据的群体评价模型与方法研究"（项目编号：71871102）和华中师范大学中央高校基本科研业务费项目"在线医疗服务环境下用户信息感知及其线下行为变化机制研究"（项目编号：CCNU17TS0009）研究成果之一

收起

An Exploratory Posts Detecting Method for MOOC Forums Based on Deep Learning

  • Dong Qingxing ,
  • Li Huayang ,
  • Cao Gaohui ,
  • Xia Lixin
Expand
  • 1 School of Information Management, Central China Normal University, Wuhan 430079;
    2 Centre for Studies of Information Resources, Wuhan University, Wuhan 430079;
    3 Tencent AI Lab, Shenzhen 518057

Received date: 2018-06-27

  Revised date: 2018-08-12

  Online published: 2019-03-05

Fold

摘要

[目的/意义]大规模在线开放课程论坛具有丰富的用户评论数据。从大量未区分的评论数据中，自动识别出知识密度较高的探索型对话并挖掘其潜在价值，对于改善教师教学质量以及提高学生知识水平具有重要影响。[方法/过程]首先利用GloVe方法训练词向量，加强对文本语义的理解，然后利用卷积神经网络自动学习文本特征，提出一种基于深度学习的探索型对话自动识别模型，并在学堂在线平台《心理学概论》课程论坛标注数据集上进行实证与对比研究。[结果/结论]实验结果显示，利用GloVe方法预训练词向量以及在训练过程中不断对词向量进行学习修正能够提高模型效果。该模型识别探索型对话的F1值为0.94，相较于传统的朴素贝叶斯方法（0.88）、逻辑斯谛回归方法（0.89）、决策树方法（0.88）以及随机森林方法（0.88）取得较大提升，具有较高的实用性和较低的学习成本。

关键词： MOOC论坛; 探索型对话; GloVe; 卷积神经网络

本文引用格式

董庆兴 , 李华阳 , 曹高辉 , 夏立新 . 基于深度学习的MOOC论坛探索型对话识别方法研究[J]. 图书情报工作, 2019 , 63(5) : 92 -99 . DOI: 10.13266/j.issn.0252-3116.2019.05.011

Abstract

[Purpose/significance] Massive Open Online Course (MOOC) forum is an important source to acquire user review data. Automatically detecting exploratory dialogues with high knowledge density from large amounts of unlabeled data and mining its potential value has a significant impact on the improvement of teaching quality and students’ mastery of knowledge. [Method/process] We proposed a new auto-detecting model based on deep learning, which firstly uses GloVe algorithm to train word embedding to reinforce semantic understanding for texts and then adopts Convolutional Neural Network (CNN) to automatically learn text features and make classifications on exploratory dialogues. An empirical and comparative study was done on the annotated dataset from the online course Introduction to Psychology on the platform of Xuetang. [Result/conclusion] Experiment result shows that using the word embedding pretrained by GloVe and fine tune it while training can improve the performance of our model. Our model gets the F1 score of 0.94, which is greatly improved compared with Naive Bayes model (0.88), Logistic Regression model (0.89), Decision Tree model (0.88) and Random Forest model (0.88) and exhibits great practicality with low learning costs.

Key words： MOOC forum; exploratory posts; GloVe; Convolutional neural network

参考文献

[1] DANIEL J. Making sense of MOOCs:Musings in a maze of myth, paradox and possibility[EB/OL].[2017-12-13]. https://jime.open.ac.uk/articles/10.5334/2012-18/.
[2] WEGERIF R, MERCER N. Computers and reasoning through talk in the classroom[J]. Language and education, 1996, 10(1):47-64.
[3] MERCER N, LITTLETON K. Dialogue and the development of children's thinking:a sociocultural approach[M]. London:Routledge, 2007.
[4] 杨现民, 唐斯斯, 李冀红. 发展教育大数据:内涵, 价值和挑战[J]. 现代远程教育研究, 2016, 139(1):50-61.
[5] SIEMENS G, LONG P. Penetrating the fog:analytics in learning and education[J]. Educause review, 2011, 46(5):31-40.
[6] CLOW D. MOOCs and the funnel of participation[C]//Proceedings of the third international conference on learning analytics and knowledge. Leuven:ACM, 2013:185-189.
[7] COETZEE D, FOX A, HEARST M A, et al. Should your MOOC forum use a reputation system?[C]//Proceedings of the 17th ACM conference on computer supported cooperative work & social computing, Baltimore, Maryland, USA:ACM, 2014:1176-1187.
[8] 陈晓美, 贯君, 王福. 虚拟社区信息运动及其规律研究[J]. 图书情报工作, 2016, 60(6):64-69.
[9] FERGUSON R. The construction of shared knowledge through asynchronous dialogue[D]. Milton Keynes:The Open University, 2009.
[10] 黄维, 赵鹏. 虚拟社区用户知识共享行为影响因素研究[J]. 情报科学, 2016, 34(4):68-73.
[11] PENNINGTON J, SOCHER R, MANNING C D. Glove:Global vectors for word representation[C]//The 2014 conference on empirical methods in natural language processing. Stroudsburg:Association for Computational Linguistics, 2014:1532-1543.
[12] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[C]//Proceedings of the IEEE. New York:IEEE, 1998:2278-2324.
[13] FERGUSON R, WEI Z, HE Y, et al. An evaluation of learning analytics to identify exploratory dialogue in online discussions[C]//Proceedings of the third international conference on learning analytics and knowledge. Leuven:ACM, 2013:85-93.
[14] SAHAMI M, DUMAIS S, HECKERMAN D, et al. A Bayesian approach to filtering junk e-mail[C]//Learning for text categorization:papers from the 1998 workshop. Palo Alto:Association for the Advancement of Artificial Intelligence, 1998:98-105.
[15] JOACHIMS T. Transductive inference for text classification using support vector machines[C]//Proceedings of the sixteenth international conference on machine learning. San Francisco:Morgan Kaufmann Publishers, 1999:200-209.
[16] 夏火松, 甄化春, 张颖烨, 等. 线上商品评论有效性分类专业领域知识模型的构建研究[J]. 情报学报, 2016, 35(9):946-954.
[17] 余本功, 李婷, 杨颖. 基于多属性加权的社会化问答社区关键词提取方法[J]. 图书情报工作, 2018, 62(5):132-139.
[18] 王东波, 何琳, 黄水清. 基于支持向量机的先秦诸子典籍自动分类研究[J]. 图书情报工作, 2017, 61(12):71-76.
[19] 夏火松, 杨培, 熊淦. 基于特征提取改进的在线评论有效性分类模型[J]. 情报学报, 2015, 34(5):493-500.
[20] 杜亚楠, 刘业政. 基于修正 G2 特征筛选的中文微博情感组合分类[J]. 情报学报, 2016, 35(4):349-357.
[21] 祁瑞华, 杨德礼, 郭旭, 等. 基于多层面文体特征的博客作者身份识别研究[J]. 情报学报, 2015, 34(6):628-634.
[22] 李晓军, 刘怀亮, 杜坤. 一种基于复杂网络模型的作者身份识别方法[J]. 图书情报工作, 2015, 59(18):102-107.
[23] BRAUD C, DENIS P. Comparing word representations for implicit discourse relation classification[C]//Proceedings of the 2015 Conference on empirical methods in natural language processing. Stroudsburg:Association for Computational Linguistics, 2015:2201-2211.
[24] TURIAN J, RATINOV L, BENGIO Y. Word representations:a simple and general method for semi-supervised learning[C]//Proceedings of the 48th annual meeting of the association for computational linguistics. Stroudsburg:Association for Computational Linguistics, 2010:384-394.
[25] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553):436-444.
[26] COLLOBERT R, WESTON J, BOTTOU L, et al. Natural language processing (almost) from scratch[J]. Journal of machine learning research, 2011, 12(8):2493-2537.
[27] KIM Y. Convolutional neural networks for sentence classification[C]//Proceedings of EMNLP. Stroudsburg:Association for Computational Linguistics, 2014.
[28] LUND K, BURGESS C. Producing high-dimensional semantic spaces from lexical co-occurrence[J]. Behavior research methods, instruments & computers, 1996, 28(2):203-208.
[29] BENGIO Y, DUCHARME R, VINCENT P, et al. A neural probabilistic language model[J]. Journal of machine learning research, 2003, 3(2):1137-1155.
[30] 安璐, 吴林. 融合主题与情感特征的突发事件微博舆情演化分析[J]. 图书情报工作, 2017, 61(15):120-129.
[31] MIKOLOV T, CHEN K, CORRADO G, et al. Efficient estimation of word representations in vector space[C]//International conference on learning representations:workshops track. La Jolla:International conference on representation Learning, 2013.
[32] KALCHBRENNER N, GREFENSTETTE E, BLUNSOM P, et al. A convolutional neural network for modelling sentences[C]//Meeting of the association for computational linguistics. Stroudsburg:Association for Computational Linguistics, 2014:655-665.
[33] SRIVASTAVA N, HINTON G E, KRIZHEVSKY A, et al. Dropout:a simple way to prevent neural networks from overfitting[J]. Journal of machine learning research, 2014, 15(1):1929-1958.
[34] KINGMA D P, BA J. Adam:a method for stochastic optimization[EB/OL].[2017-12-30]. https://arxiv.org/abs/1412.6980.
[35] 唐晓波, 罗颖利. 融入情感差异和用户兴趣的微博转发预测[J]. 图书情报工作, 2017, 61(9):102-110.
Options
文章导航

/

〈 〉