دكتوراه
العلوم والتقنية
Durham University
مجال التميز | تميز دراسي وبحثي |
البحوث المنشورة | |
البحث (1): | |
عنوان البحث: | Towards Designing Profitable Courses: Predicting Student Purchasing Behaviour in MOOCs |
رابط إلى البحث: | https://link.springer.com/content/pdf/10.1007/s40593-021-00246-2.pdf |
تاريخ النشر: | 23/03/2021 |
موجز عن البحث: | Since their ‘official’ emergence in 2012 (Gardner and Brooks 2018), massive open online courses (MOOCs) have been growing rapidly. They offer low-cost education for both students and content providers; however, currently there is a very low level of course purchasing (less than 1% of the total number of enrolled students on a given online course opt to purchase its certificate). The most recent literature on MOOCs focuses on identifying factors that contribute to student success, completion level and engagement. One of the MOOC platforms’ ultimate targets is to become self-sustaining, enabling partners to create revenues and offset operating costs. Nevertheless, analysing learners’ purchasing behaviour on MOOCs remains limited. Thus, this study aims to predict students purchasing behaviour and therefore a MOOCs revenue, based on the rich array of activity clickstream and demographic data from learners. Specifically, we compare how several machine learning algorithms, namely RandomForest, GradientBoosting, AdaBoost and XGBoost can predict course purchasability using a large-scale data collection of 23 runs spread over 5 courses delivered by The University of Warwick between 2013 and 2017 via FutureLearn. We further identify the common representative predictive attributes that influence a learner’s certificate purchasing decisions. Our proposed model achieved promising accuracies, between 0.82 and 0.91, using only the time spent on each step. We further reached higher accuracy of 0.83 to 0.95, adding learner demographics (e.g. gender, age group, level of education, and country) which showed a considerable impact on the model’s performance. The outcomes of this study are expected to help design future courses and predict the profitability of future runs; it may also help determine what personalisation features could be provided to increase MOOC revenue. |
المؤتمرات العلمية | |
المؤتمر (1): | |
عنوان المؤتمر: | ITS2021 :17th International Conference on Intelligent Tutoring Systems |
تاريخ الإنعقاد: | 07-11/06/2021 |
مكان الإنعقاد: | Athens, Greece (virtual) |
طبيعة المشاركة: | Paper presentation |
عنوان المشاركة: | Predicting Certification in MOOCs based on Students’ Weekly Activities |
ملخص المشاركة: | Massive Open Online Courses (MOOCs) have been growing rapidly, offering low-cost knowledge for both learners and content providers. However, currently there is a very low level of course purchasing (less than 1% of the total number of enrolled students on a given online course opt to purchase its certificate). This can impact seriously the business model of MOOCs. Nevertheless, MOOC research on learners’ purchasing behaviour on MOOCs remains limited. Thus, the umbrella question that this work tackles is if learner’s data can predict their purchasing decision (certification). Our fine-grained analysis attempts to uncover the latent correlation between learner activities and their decision to purchase. We used a relatively large dataset of 5 courses of 23 runs obtained from the less studied MOOC platform of FutureLearn to: (1) statistically compare the activities of non-paying learners with course purchasers, (2) predict course certification using different classifiers, optimising for this naturally strongly imbalanced dataset. Our results show that learner activities are good predictors of course purchasibility; still, the main challenge was that of early prediction. Using only student number of step accesses, attempts, correct and wrong answers, our model achieve promising accuracies, ranging between 0.81 and 0.95 across the five courses. The outcomes of this study are expected to help design future courses and predict the profitability of future runs; it may also help determine what personalisation features could be provided to increase MOOC revenue. |
الرابط: | https://its2021.iis-international.org/full-conference-program/ |
المؤتمر (2): | |
عنوان المؤتمر: | ITS 2019: 15th International Conference on Intelligent Tutoring Systems |
تاريخ الإنعقاد: | June 3–7, 2019 |
مكان الإنعقاد: | Kingston, Jamaica |
طبيعة المشاركة: | Paper presentation |
عنوان المشاركة: | Predicting MOOCs Dropout Using only two easily obtainable Features from the First Week’s Activities |
ملخص المشاركة: | While Massive Open Online Course (MOOCs) platforms provide knowledge in a new and unique way, the very high number of dropouts is a significant drawback. Several features are considered to contribute towards learner attrition or lack of interest, which may lead to disengagement or total dropout. The jury is still out on which factors are the most appropriate predictors. However, the literature agrees that early prediction is vital to allow for a timely intervention. Whilst feature-rich predictors may have the best chance for high accuracy, they may be unwieldy. This study aims to predict learner dropout early-on, from the first week, by comparing several machinelearning approaches, including Random Forest, Adaptive Boost, XGBoost and GradientBoost Classifiers. The results show promising accuracies (82% – 94%) using as little as 2 features. We show that the accuracies obtained outperform state of the art approaches, even when the latter deploy several features. |
الرابط: | https://its2019.iis-international.org/wp-content/uploads/ITS2019-Program.pdf |
المؤتمر (3): | |
عنوان المؤتمر: | The 3rd International Conference on Information and Education Innovations |
تاريخ الإنعقاد: | 05/06/2018 |
مكان الإنعقاد: | London, UK |
طبيعة المشاركة: | Paper presentation |
عنوان المشاركة: | On the Need for Fine-Grained Analysis of Gender Versus Commenting Behaviour in MOOCs |
ملخص المشاركة: | Stereotyping is the first type of adaptation ever proposed. However, the early systems have never dealt with the numbers of learners that current Massive Open Online Courses (MOOCs) provide. Thus, the umbrella question that this work tackles is if learner characteristics can predict their overall, but also fine-grain behaviour. Earlier results point at differences related to gender or to age. Here, we analyse gender versus commenting behaviour. Our fine-grained analysis shows that the result may further depend on the course topic, or even week. Surprisingly, for instance, women chat less in a Psychology-related course, but more (or similar) on a Computer Science course. These results are analysed in this paper in details, including two different methods of averaging comments, leading to remarkably different results. The outcomes can help in informing future runs, in terms of potential personalised feedback for teachers and students. |
الرابط: | https://dl.acm.org/doi/10.1145/3234825.3234833 |
المؤتمر (4): | |
عنوان المؤتمر: | 27th International Conference on Information Systems Development |
تاريخ الإنعقاد: | 22-24/08/2018 |
مكان الإنعقاد: | Lund, Sweden |
طبيعة المشاركة: | Paper presentation |
عنوان المشاركة: | How is learning fluctuating? FutureLearn MOOCs fine-grained temporal analysis and feedback to teachers |
ملخص المشاركة: | Data-intensive analysis of massive open online courses (MOOCs) is popular. Researchers have been proposing various parameters conducive to analysis and prediction of student behaviour and outcomes in MOOCs, as well as different methods to analyse and use these parameters, ranging from statistics, to NLP, to ML, and even graph analysis. In this paper, we focus on patterns to be extracted, and apply systematic data analysis methods in one of the few genuinely large-scale data collection of 5 MOOCs, spread over 21 runs, on FutureLearn, a UK-based MOOCs provider, that, whilst offering a broad range of courses from many universities, NGOs and other institutions, has been less evaluated, in comparison to, e.g., its American counterparts. We analyse temporal quiz solving patterns; specifically, the less explored issue on how the first number of weeks of data predicts activities in the last weeks; we also address the classical MOOC question on the completion chance. Finally, we discuss the type of feedback a teacher or designer could receive on their MOOCs, in terms of fine-grained analysis of their material, and what personalisation could be provided to a student. |