LA has emerged with EBD. LA “refers to the analysis and interpretation of data related to learners’ behaviors and interactions during the learning process, as well as learners’ profiles and the learning contexts they are situated in” (Hwang, Chu, & Yin, 2017, p.143). Before the emergence of LA, most research mainly concentrated on the analysis of learning outcomes, with considerable limitations in terms of data and sample sizes. LA focuses on extracting and analyzing meaningful information from EBD and as the data size is massive big data analytics tools must be used. LA helps to discover novel and potentially useful information in large amounts of unstructured data. With the LA results, teachers and students can change their teaching/learning strategies. With the new strategies, new EBD will be generated and new LA results can continuously be provided to teachers and students.
Previous research has argued that LA can facilitate the design of learning systems, educational materials and activities to improve educational effectiveness and optimize learning environments (Sutcliffe & Hart, 2016; Law & Larusdottir, 2015; Brajnik & Gabrielli, 2010; Greller & Drachsler, 2012). Romero and Ventura (2010) claim that each stakeholder obtains different benefits from LA:
• a. With regards to learners, they can benefit from personalization, recommendations of resources and activities, an improved learning experience and adaptive clues;
• b. Teachers can use LA to receive feedback, to examine both the learning and the behavior of the learners, to identify the students who need support, to determine which mistakes occur more often and to improve the effectiveness of some activities;
• c. Course developers can benefit from an evaluation of the courses’ structure and its impact for learning, to assess course materials, to identify the most valuable data mining methods according to different tasks and to develop learning models.
• d.With concern to administrators of educational institutions, they can use LA to organize resources, to improve their offer of educational programs and to assess both teachers and the effectiveness of curricula.
Considering these benefits of LA, many educational institutions have begun to collect EBD to perform LA and to incorporate it in their curricula. For example, the world’s first graduate program in LA was created at Teachers College of Columbia University (USA). A similar program was also created at the University of Pennsylvania (USA), named Learning Science & Technologies. Moreover, analogous programs also appeared in other universities, such as Northeastern University (USA), Boston University (USA), Carnegie Mellon University (USA), University of California Berkeley (USA), Georgetown University (USA), University of Sydney (Australian), University of Edinburgh (England). Some LA research groups were equally created: the University of Maryland (USA) created the LA Research Group; and Kyusyu University (Japan) created the LA Center.
With the growing popularity of LA and EBD as research subjects, the focus has been placed on prediction, structure discovery, and relationship mining. Research within these areas has been conducted at different levels, such as at course or degree level (Asif et al., 2017). In terms of prediction, a central research topic is predicting students’ educational outcomes. In structure discovery, the emphasis is on finding structure, patterns and data points in a set of data without any ground truth or a priori idea of what should be found (Baker & Inventado, 2014). Relationship mining involves discovering relationships between variables in a dataset. These relationships are seen as rules of data for later use (Bousbia & Belamri, 2013).
Despite the fact that there is a significant body of research in LA and EBD, there are still numerous challenges and questions that remain to be addressed: how can EBD be built and used by integrating different kinds of data from online learning systems such as learning management systems and game based learning systems; how to measure the adequateness and effectiveness of the LA results; how to promote the use of online learning systems for collecting EBD; how to protect personal privacy when collecting data from online learning systems, including privacy and security control policies; how to employ LA results to support learning designs; how to integrate learning theories and strategies with LA; how to employ LA approaches in various application domains.
1. the extended version of the conference papers should have at least 45% difference from the conference version and the difference should be reflected in the analysis, design, system, experiment/evaluation, data collection and analysis, and findings and discussions;
2. the paper should be NOT including too theoretical contents (e.g., math, symbols and equations) and the paper needs to have running examples with simulated or real datasets to explain correspondent equations, definitions, atoms, theorems, and proofs if those theoretical content is extremely important for readers to understand their research;
3. the paper should be strongly connected to the practical usages in distance education or e-learning, which means, a research is used for people to learn in the classroom (or with others at same place together at the same time) would never be accepted;
4. the paper needs to have experiment or evaluation with proper quantitative and/or qualitative data analysis, findings and discussions;