Discovering Learners Behaviour Patterns From Log Files Using LSA

Discovering Learners Behaviour Patterns From Log Files Using LSA

Iness Nedji Milat (Badji Mokhtar University, Annaba, Algeria), Hassina Seridi (LABGED Laboratory, Badji Mokhtar University, Annaba, Algeria) and Abdelkader Moudjari (MISC Laboratory, Abdelhamid Mehri University, Constantine, Algeria)
Copyright: © 2020 |Pages: 24
DOI: 10.4018/IJDET.2020040106
OnDemand PDF Download:
No Current Special Offers


Recently, discovering learner behaviour has taken more attention in the field of e-learning. It aims to gain useful insights into the learning process of students despite the absence of direct interaction with teachers. In fact, the only available source of information in such environments is the log file that represents all possible interactions of learners with the e-learning system. This log file is characterised by the presence of noise, incomplete information, and a huge amount of data. In this article, a new approach based on learner trails analysis from the log file is proposed. It aims to discover the patterns of the real behaviour of learners and to determine their pedagogic orientations. The latent semantic analysis (LSA) method is used to extract the relationship between learners who have the same behaviour and to overcome the noise problem. The proposed approach has been validated using synthetic and genuine log files. The obtained results show the efficiency of the proposed method of discovering the behaviours of learners.
Article Preview


In the context of e-learning systems, instructors are not confronted with their learners to supervise them. They do not have enough information about what are learners doing, and whether they have problems during the learning process or not. Consequently, instructors cannot take the appropriate decisions in time to help learners (Graf & Liu, 2010; Martinez et al., 2015; Pardo, 2014).

To deal with these issues, several works have been proposed to analyse learner behaviours. This analysis aims to acquire knowledge about the learning process and to understand the learner behaviours in the e-learning system. This knowledge can be used by different stakeholders such as learners, instructors, designers and pedagogues. For example, learners could get self-awareness about their learning progress. The Instructors could supervise their learners. Additionally, course designers and/or pedagogues can adapt and personalise the learning process (Bourguignon, 2007; Romero et al., 2013; Pardo, 2014; Mahajan, 2014; Martínez et al., 2015; Cereso et al., 2016).

Actually, the analysis of learner interactions serves as a mechanism to understand their behaviours in the e-learning system. This analysis is achieved by using log files-based processing (Bousbia et al., 2010). Log files save all events that can occur during the learning process. However, these events could not be exploited directly by different stakeholders because they are low-level, noisy, incomplete, and huge data. For this, it is difficult to extract knowledge directly from these log files.

Among the techniques used in the field of e-learning, to extract useful knowledge from log files, the authors find Educational Data Mining (EDM) (Neuhauser, 2002; Romero & Ventura, 2010; Siemens & Long, 2011; Agosti et al., 2012; Greller & Drachsler, 2012; Stater et al., 2016; Bakhshinategh et al., 2018). This technique is used to well-understand learner interactions and to discover their behaviours (Cereso et al., 2016). EDM focuses on developing and applying computerised methods to detect patterns in large collections of educational data, which would otherwise be difficult or even impossible to analyse (Romero & Ventura, 2010; Peña-Ayala, 2013). These methods include data mining, machine learning, statistics, data visualization, and modelling methods (Baker & Siemens, 2014; Bakhshinategh et al., 2018).

Complete Article List

Search this Journal:
Volume 20: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 19: 4 Issues (2021)
Volume 18: 4 Issues (2020)
Volume 17: 4 Issues (2019)
Volume 16: 4 Issues (2018)
Volume 15: 4 Issues (2017)
Volume 14: 4 Issues (2016)
Volume 13: 4 Issues (2015)
Volume 12: 4 Issues (2014)
Volume 11: 4 Issues (2013)
Volume 10: 4 Issues (2012)
Volume 9: 4 Issues (2011)
Volume 8: 4 Issues (2010)
Volume 7: 4 Issues (2009)
Volume 6: 4 Issues (2008)
Volume 5: 4 Issues (2007)
Volume 4: 4 Issues (2006)
Volume 3: 4 Issues (2005)
Volume 2: 4 Issues (2004)
Volume 1: 4 Issues (2003)
View Complete Journal Contents Listing