Temporal emotion-aspect modeling for discovering what students are concerned about in online course forums

ABSTRACT Textual data, as a key carrier of learning feedback, is continuously produced by many students within course forums. The temporal nature of discussion requires students’ emotions and concerned aspects (e.g. teaching styles, learning activities, etc.) to be dynamically tracked for understanding learning requirements. To characterize dynamics of emotion-aspects, this paper presents an unsupervised model, namely temporal emotion-aspect model (TEAM), modeling time jointly with emotions and aspects to capture emotion-aspect evolutions over time. Especially, the primary output of TEAM is two-fold: emotion-specific aspect probabilistic distributions and their evolutions over time. With the discussion data from two online course forums, we validated the performance of TEAM and utilized this model to discover most concerned emotion-aspects as well as their evolutionary trends for the whole learning group and different achievement levels of groups, respectively. The results indicated that, (1) content-related aspects were the main focus with higher probabilities to negative and confused emotions; (2) there were higher probabilities of emotional expressions in the initial and final stages of a semester; (3) compared with high- and medium-achieving students, low-achieving students were less active in emotional engagement on the whole, and tended to express more confusion in the final stage of a semester.

[1]  James R. Foulds,et al.  Weakly Supervised Models of Aspect-Sentiment for Online Course Discussion Forums , 2015, ACL.

[2]  Jiajia Wang,et al.  Sentiment contagion in complex networks , 2014 .

[3]  Li Li,et al.  Analyzing Topic-Sentiment and Topic Evolution over Time from Social Media , 2016, KSEM.

[4]  Alyssa Friend Wise,et al.  Topic models to support instructors in MOOC forums , 2017, LAK.

[5]  Pythagoras Karampiperis,et al.  Sentiment Analysis: A tool for Rating Attribution to Content in Recommender Systems , 2012, RecSysTEL@EC-TEL.

[6]  Alan F. Smeaton,et al.  Topic-dependent sentiment analysis of financial blogs , 2009, TSA@CIKM.

[7]  Zhi Liu,et al.  Emotion and Associated Topic Detection for Course Comments in a MOOC Platform , 2016, 2016 International Conference on Educational Innovation through Technology (EITT).

[8]  Laurie P. Dringus,et al.  Using data mining as a strategy for assessing asynchronous discussion forums , 2005, Comput. Educ..

[9]  Kalina Yacef,et al.  Analysis of collaborative writing processes using revision maps and probabilistic topic models , 2013, LAK '13.

[10]  Haji Binali,et al.  A new significant area: Emotion detection in E-learning using opinion mining techniques , 2009, 2009 3rd IEEE International Conference on Digital Ecosystems and Technologies.

[11]  A. Graesser,et al.  Dynamics of affective states during complex learning , 2012 .

[12]  Erin Shaw,et al.  Discerning Affect in Student Discussions , 2010 .

[13]  James C. Lester,et al.  Evaluating the consequences of affective feedback in intelligent tutoring systems , 2009, 2009 3rd International Conference on Affective Computing and Intelligent Interaction and Workshops.

[14]  Sintija Petrovica,et al.  Adaptation of tutoring to students' emotions in emotionally intelligent tutoring systems , 2013, 2013 Second International Conference on E-Learning and E-Technologies in Education (ICEEE).

[15]  Yulan He,et al.  Joint sentiment/topic model for sentiment analysis , 2009, CIKM.

[16]  T. L. Kelley The selection of upper and lower groups for the validation of test items. , 1939 .

[17]  Wei Gao,et al.  Dynamic joint sentiment-topic model , 2013, ACM Trans. Intell. Syst. Technol..

[18]  Noémie Elhadad,et al.  An Unsupervised Aspect-Sentiment Model for Online Reviews , 2010, NAACL.

[19]  Jon Atle Gulla,et al.  Retraction Note to: A joint model for analyzing topic and sentiment dynamics from large-scale online news , 2018, World Wide Web.

[20]  Lise Getoor,et al.  Understanding MOOC Discussion Forums using Seeded LDA , 2014, BEA@ACL.

[21]  Sannyuya Liu,et al.  Joint Exploration of Negative Academic Emotion and Topics in Student-Generated Online Course Comments , 2017, 2017 International Conference of Educational Innovation through Technology (EITT).

[22]  Zied Kechaou,et al.  Improving e-learning with sentiment analysis of users' opinions , 2011, 2011 IEEE Global Engineering Education Conference (EDUCON).

[23]  K. Takano,et al.  Self-rumination, self-reflection, and depression: self-rumination counteracts the adaptive effect of self-reflection. , 2009, Behaviour research and therapy.

[24]  Zhenming Liu,et al.  Learning about Social Learning in MOOCs: From Statistical Analysis to Generative Model , 2013, IEEE Transactions on Learning Technologies.

[25]  Kristy Elizabeth Boyer,et al.  Unsupervised modeling for understanding MOOC discussion forums: a learning analytics approach , 2015, LAK.

[26]  A. Graesser,et al.  Confusion can be beneficial for learning. , 2014 .

[27]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[28]  Alice H. Oh,et al.  Aspect and sentiment unification model for online review analysis , 2011, WSDM '11.

[29]  Kun Yang,et al.  Dynamic non-parametric joint sentiment topic mixture model , 2015, Knowl. Based Syst..

[30]  Jon Atle Gulla,et al.  RETRACTED ARTICLE: A joint model for analyzing topic and sentiment dynamics from large-scale online news , 2017, World Wide Web.

[31]  Hongfei Lin,et al.  A Convolution-LSTM-Based Deep Neural Network for Cross-Domain MOOC Forum Post Classification , 2017, Inf..

[32]  Carolyn Penstein Rosé,et al.  Exploring the Effect of Student Confusion in Massive Open Online Courses , 2016, EDM.

[33]  Hongfei Yan,et al.  Jointly Modeling Aspects and Opinions with a MaxEnt-LDA Hybrid , 2010, EMNLP.

[34]  Carolyn Penstein Rosé,et al.  Sentiment Analysis in MOOC Discussion Forums: What does it tell us? , 2014, EDM.

[35]  Pilar Rodríguez Marín,et al.  Extracting Emotions from Texts in E-Learning Environments , 2012, 2012 Sixth International Conference on Complex, Intelligent, and Software Intensive Systems.

[36]  Sannyuya Liu,et al.  Exploring Students’ Engagement Patterns in SPOC Forums and their Association with Course Performance , 2018 .

[37]  Xianghua Fu,et al.  Multi-aspect sentiment analysis for Chinese online social reviews based on topic modeling and HowNet lexicon , 2013, Knowl. Based Syst..

[38]  Sergio Salmeron-Majadas,et al.  Towards Enriching an ITS with Affective Support , 2013, UMAP Workshops.

[39]  Stefan M. Rüger,et al.  Weakly Supervised Joint Sentiment-Topic Detection from Text , 2012, IEEE Transactions on Knowledge and Data Engineering.

[40]  Samuel E. Ehrenreich,et al.  The BlackBerry project: capturing the content of adolescents' text messaging. , 2012, Developmental psychology.

[41]  Jian Zhao,et al.  Visual Analysis of MOOC Forums with iForum , 2017, IEEE Transactions on Visualization and Computer Graphics.

[42]  Xiaohong Guan,et al.  Modelling multi-topic information propagation in online social networks based on resource competition , 2017, J. Inf. Sci..

[43]  D. C. Howell Statistical Methods for Psychology , 1987 .

[44]  Gokhan Tur,et al.  LDA Based Similarity Modeling for Question Answering , 2010, HLT-NAACL 2010.

[45]  Robert L. Mercer,et al.  An Estimate of an Upper Bound for the Entropy of English , 1992, CL.

[46]  Qun Liu,et al.  Chinese Lexical Analysis Using Hierarchical Hidden Markov Model , 2003, SIGHAN.

[47]  Koji Eguchi,et al.  Sentiment Retrieval using Generative Models , 2006, EMNLP.

[48]  Charles Elkan,et al.  Clustering documents with an exponential-family approximation of the Dirichlet compound multinomial distribution , 2006, ICML.

[49]  E. Hovy,et al.  Mining and Assessing Discussions on the Web through Speech Act Analysis , 2006 .

[50]  Panagiotis Adamopoulos,et al.  What makes a great MOOC? An interdisciplinary analysis of student retention in online courses , 2013, ICIS.

[51]  Ryan Shaun Joazeiro de Baker,et al.  Sequences of Frustration and Confusion, and Learning , 2013, EDM.

[52]  Jihie Kim,et al.  Sentiment Analysis of a Student Q&A Board for Computer Science , 2009 .

[53]  Dan Wang,et al.  Using a Learner-Topic Model for Mining Learner Interests in Open Learning Environments , 2018, J. Educ. Technol. Soc..

[54]  Mihaela Cocea,et al.  Predicting Students' Emotions Using Machine Learning Techniques , 2015, AIED.

[55]  Sabine Loudcher,et al.  A Joint Model for Topic-Sentiment Evolution over Time , 2014, 2014 IEEE International Conference on Data Mining.

[56]  K. VanLehn,et al.  Why Do Only Some Events Cause Learning During Human Tutoring? , 2003 .

[57]  Yew Haur Lee,et al.  Mining sentiments in SMS texts for teaching evaluation , 2012, Expert Syst. Appl..

[58]  XianghuaFu,et al.  Multi-aspect sentiment analysis for Chinese online social reviews based on topic modeling and HowNet lexicon , 2013 .

[59]  Lise Getoor,et al.  Modeling Learner Engagement in MOOCs using Probabilistic Soft Logic , 2013 .

[60]  Longzhuang Li,et al.  Recognizing and regulating e-learners' emotions based on interactive Chinese texts in e-learning systems , 2014, Knowl. Based Syst..

[61]  Carolyn Penstein Rosé,et al.  Exploring the Effect of Confusion in Discussion Forums of Massive Open Online Courses , 2015, L@S.

[62]  Lori Lockyer,et al.  Inside Out , 2017 .

[63]  SangKeun Lee,et al.  Joint multi-grain topic sentiment: modeling semantic aspects for online reviews , 2016, Inf. Sci..

[64]  Bikram Sengupta,et al.  Student Emotion, Co-occurrence, and Dropout in a MOOC Context , 2016, EDM.

[65]  V. Shute Focus on Formative Feedback , 2007 .

[66]  Martin Ester,et al.  On the design of LDA models for aspect-based opinion mining , 2012, CIKM.