Human-Centered Reflective Writing Analytics with a Deep Transfer Learning Approach

Ming Liu,Adithya Krishna,Simon Buckingham Shum

DOI: https://doi.org/10.2139/ssrn.3991773

2021-01-01

Abstract:Download This Paper Open PDF in Browser Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Human-Centered Reflective Writing Analytics with a Deep Transfer Learning Approach 26 Pages Posted: 22 Dec 2021 See all articles by Ming LiuMing LiuSouthwest UniversityAdithya Krishnaaffiliation not provided to SSRNSimon Buckingham Shumaffiliation not provided to SSRN Abstract Reflective writing analytics (RWA) has great potential to support experiential learning by providing accurate and actionable feedback. However, current RWA research focused on the development of an accurate analytical model and ignored the importance of the actionable feedback generation. Thus, this article presents two studies. In the first study, we introduce a multi-stage co-design model which learning analytics designers and educators work together to transform the reflective writing analytics to actionable feedback. Over the past two years, we have successfully integrated the reflective analytics tool into 5 different university subjects. Our student survey results illustrated that most students have perceived the system feedback is useful for them to understand the reflective writing criteria, identified the weakness of their writings and build confidence in completing the task. However, the limitation of the current rule-based reflective element classification performance affect students' negative perception of the system feedback. Moreover, more potential educational requirements, such as link-to-theory and beliefs reflective elements, were identified during the co-design process. Thus, in the second study, we developed a human annotated reflective writing dataset and proposed a deep transfer learning approach for improving reflective element classification. Our study results suggested that the pre-trained Bidirectional Encoder Representations from Transformers model combining with linguistic textual features can address the small and imbalanced dataset issues that outperformed traditional random forest model by 20% accuracy. Keywords: Reflective Writing Analytics, Feedback Generation, Human Centered Learning Analytics Suggested Citation: Suggested Citation Liu, Ming and Krishna, Adithya and Buckingham Shum, Simon, Human-Centered Reflective Writing Analytics with a Deep Transfer Learning Approach. Available at SSRN: https://ssrn.com/abstract=3991773 Ming Liu (Contact Author) Southwest University ( email ) College of Economics and ManagementChongqing, 400715China Adithya Krishna affiliation not provided to SSRN ( email ) No Address Available Simon Buckingham Shum affiliation not provided to SSRN Download This Paper Open PDF in Browser Do you have a job opening that you would like to promote on SSRN? Place job opening Paper statistics Downloads 0 Abstract Views 1 PlumX Metrics Feedback Feedback to SSRN Feedback (required) Email (required) Submit If you need immediate assistance, call 877-SSRNHelp (877 777 6435) in the United States, or +1 212 448 2500 outside of the United States, 8:30AM to 6:00PM U.S. Eastern, Monday - Friday. Submit a Paper Section 508 Text Only Pages SSRN Quick Links SSRN Solutions Research Paper Series Conference Papers Partners in Publishing Jobs & Announcements Newsletter Sign Up SSRN Rankings Top Papers Top Authors Top Organizations About SSRN SSRN Objectives Network Directors Presidential Letter Announcements Contact us FAQs Copyright Terms and Conditions Privacy Policy We use cookies to help provide and enhance our service and tailor content. To learn more, visit Cookie Settings. This page was processed by aws-apollo1 in 0.203 seconds