E-book user modelling through learning analytics: the case of learner engagement and reading styles. Issue 5 (18th August 2019)

• Record Type: Journal Article
• Authors: Boticki, Ivica; Akçapınar, Gökhan; Ogata, Hiroaki
• Is Part Of: Interactive learning environments. Volume 27:Issue 5/6(2019)
• Page Start: 754
• Page End: 765
• Publication Date: 2019-08-18
• Subjects: User modelling -- engagement -- reading styles -- learning analytics -- e-books; Educational technology -- United States -- Periodicals; 371.33
• DOI: 10.1080/10494820.2019.1610459
• Languages: English
• ISSNs: 1049-4820

Abstract

ABSTRACT: In this paper log data on e-book usage is used as part of a learning analytics approach to generate user models which describe university students' characteristics in multiple dimensions. E-book usage is logged and analysed to extract information on how users use e-books for academic purposes. Two cases contributing to user modelling are presented: generating user engagement and user reading style variables describing students' e-book usage. The paper discusses the approach to the variables extraction from data logs, discusses the relationship between the variables and proposes a user modelling approach specifically targeted at building high-level services related to educational e-books. The study examines how can user model variables from different dimensions be used in getting a higher-level overview of user performance and skills; and proposes services which leverage the variables of learner engagement and reading styles.