Towards better affect detectors: effect of missing skills, class features and common wrong answers

The well-studied Baker et al., affect detectors on boredom, frustration, confusion and engagement concentration with ASSISTments dataset were used to predict state tests scores, college enrollment, and even whether a student majored in a STEM field. In this paper, we present three attempts to improve upon current affect detectors. The first attempt analyzed the effect of missing skill tags in the dataset to the accuracy of the affect detectors. The results show a small improvement after correctly tagging the missing skill values. The second attempt added four features related to student classes for feature selection. The third attempt added two features that described information about student common wrong answers for feature selection. Result showed that two out of the four detectors were improved by adding the new features.

[1]  Jacob Cohen.  A Coefficient of Agreement for Nominal Scales , 1960 .

[2]  Yutao Wang,et al.  Class vs. Student in a Bayesian Network Student Model , 2013, AIED.

[3]  Scotty D. Craig,et al.  Affect and learning: An exploratory look into the role of affect in learning with AutoTutor , 2004 .

[4]  Neil T. Heffernan,et al.  Which Is More Responsible for Boredom in Intelligent Tutoring Systems: Students (Trait) or Problems (State)? , 2013, 2013 Humaine Association Conference on Affective Computing and Intelligent Interaction.

[5]  Zachary A. Pardos,et al.  Affective states and state tests: investigating how affect throughout the school year predicts end of year learning outcomes , 2013, LAK '13.

[6]  Neil T. Heffernan,et al.  Predicting College Enrollment from Student Interaction with an Intelligent Tutoring System in Middle School , 2013, EDM.

[7]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[8]  Arthur C. Graesser,et al.  Automatic detection of learner’s affect from conversational cues , 2008, User Modeling and User-Adapted Interaction.

[9]  Neil T. Heffernan,et al.  Population validity for educational data mining models: A case study in affect detection , 2014, Br. J. Educ. Technol..

[10]  Ingo Mierswa,et al.  YALE: rapid prototyping for complex data mining tasks , 2006, KDD '06.

[11]  Ryan Shaun Joazeiro de Baker,et al.  Towards Predicting Future Transfer of Learning , 2011, AIED.

[12]  James C. Lester,et al.  Modeling Learner Affect with Theoretically Grounded Dynamic Bayesian Networks , 2011, ACII.