Search

Conversational Intelligent Tutoring Systems (CITS) are agent based e-learning systems which deliver tutorial content through discussion, asking and answering questions, identifying gaps in knowledge and providing feedback in natural language. Personalisation and adaptation for CITS are current research focuses in the field. Classroom studies have shown that experienced human tutors automatically, through experience, estimate a learner’s level of subject comprehension during interactions and modify lesson content, activities and pedagogy in response. This paper introduces Hendrix 2.0, a novel CITS capable of classifying e-learner comprehension in real-time from webcam images. Hendrix 2.0 integrates a novel image processing and machine learning algorithm, COMPASS, that rapidly detects a broad range of non-verbal behaviours, producing a time-series of comprehension estimates on a scale from -1.0 to +1.0. This paper reports an empirical study of comprehension classification accuracy, during which 51 students at Manchester Metropolitan University undertook conversational tutoring with Hendrix 2.0. The authors evaluate the accuracy of strong comprehension and strong non-comprehension classifications, during conversational questioning. The results show that the COMPASS comprehension classifier achieved normalised classification accuracy of 75%.

Conversational Intelligent Tutoring Systems (CITS) are agent based e-learning systems which deliver tutorial content through discussion, asking and answering questions, identifying gaps in knowledge and providing feedback in natural language. Personalisation and adaptation for CITS are current research focuses in the field. Classroom studies have shown that experienced human tutors automatically, through experience, estimate a learner’s level of subject comprehension during interactions and modify lesson content, activities and pedagogy in response. This paper introduces Hendrix 2.0, a novel CITS capable of classifying e-learner comprehension in real-time from webcam images. Hendrix 2.0 integrates a novel image processing and machine learning algorithm, COMPASS, that rapidly detects a broad range of non-verbal behaviours, producing a time-series of comprehension estimates on a scale from -1.0 to +1.0. This paper reports an empirical study of comprehension classification accuracy, during which 51 students at Manchester Metropolitan University undertook conversational tutoring with Hendrix 2.0. The authors evaluate the accuracy of strong comprehension and strong non-comprehension classifications, during conversational questioning. The results show that the COMPASS comprehension classifier achieved normalised classification accuracy of 75%.

Conversational Intelligent Tutoring Systems (CITS) that automatically adapt to learning styles (LS) can improve learning, however current modelling of LS has ignored Neutral learners. This paper presents research examining the ability of data mining algorithms to predict LS dimensions from behaviour captured during natural language tutorials with Oscar CITS. Two datasets, 2ClassBDS and 3ClassBDS, were cleaned and prepared for the data mining task of predicting student LS. Each dataset comprised four sub-datasets representing the four Felder-Silverman LS dimensions. 3ClassBDS included a third Neutral class describing individuals with a balance of LS preferences. Naïve Bayes, Decision Trees, Lazy Learning and Neural Networks algorithms were applied to each dataset and parameters adjusted to improve prediction accuracies. The 2ClassBDS dataset results show good prediction, with decision trees (Simple CART) achieving accuracies of 81.33-86.66%. For 3ClassBDS results were mixed, with the J48 algorithm achieving 56-73% accuracy, indicating that further work and data is needed.

Conversational Intelligent Tutoring Systems (CITS) that automatically adapt to learning styles (LS) can improve learning, however current modelling of LS has ignored Neutral learners. This paper presents research examining the ability of data mining algorithms to predict LS dimensions from behaviour captured during natural language tutorials with Oscar CITS. Two datasets, 2ClassBDS and 3ClassBDS, were cleaned and prepared for the data mining task of predicting student LS. Each dataset comprised four sub-datasets representing the four Felder-Silverman LS dimensions. 3ClassBDS included a third Neutral class describing individuals with a balance of LS preferences. Naïve Bayes, Decision Trees, Lazy Learning and Neural Networks algorithms were applied to each dataset and parameters adjusted to improve prediction accuracies. The 2ClassBDS dataset results show good prediction, with decision trees (Simple CART) achieving accuracies of 81.33-86.66%. For 3ClassBDS results were mixed, with the J48 algorithm achieving 56-73% accuracy, indicating that further work and data is needed.

This paper describes an adaptive online conversational intelligent tu-toring system (CITS) called Oscar that delivers a personalised natural language tutorial. During the tutoring conversation, Oscar CITS dynamically predicts and adapts to a student’s learning style. Oscar CITS aims to mimic a human tutor by using knowledge of learning styles to adapt its tutoring style and improve the effectiveness of the learning experience. Learners can intuitively explore and discuss topics in natural language, helping to establish a deeper understanding of the topic and boost confidence. An initial study into the adaptation to learn-ing styles is reported which produced encouraging results and positive test score improvements. The results show that students experiencing a tutorial adapted to suit their learning styles performed significantly better than those experiencing an unsuited tutorial.