Article Preview
TopIntroduction
The last few decades have witnessed major developments in information and communications technologies, resulting in a significant increase the amount of “big data” (Tulasi, 2013). The big data (BD) generated from integrated technologies and systems shared by different organizations has also increased (Kaisler et al., 2013). This is due to applying artificial intelligence (AI) and Internet of things (IoT) technologies within systems across different sectors. Researchers’ attention is now on how to exploit and implement BD to enable organizations to acquire new knowledge in response to emerging opportunities and challenges (Sin & Muthu, 2015).
Education institutions seeking to develop and to progress must employ big data analytics (BDA) to exploit creative opportunities and to access and develop ideas. The process of mining education data is typically referred to as educational data mining (EDM; Sin & Muthu, 2015). According to the International Educational Data Mining Society, EDM is defined as “an emerging field targeting the development of methods to explore data sets unique to the education settings” (International Educational Data Mining Society, 2011, p.601). As such, it is an interdisciplinary field that relies on the application of machine learning, data mining (DM), statistics, recommender systems, psycho-pedagogy, and information retrieval to large volumes of educational data.
Higher education sectors (HESs) play a vital role in a nation's overall social and economic development. In turn, several factors contribute to establishing a quality HES, including goal-based processes, curriculum relevance in terms of discipline-specific subjects to meet business and industry needs, and the effective delivery of teaching and learning activities (Rajni & Malaya, 2015). HES are also recognized as an important contributor to technology-driven social advancement (Al-Shaya et al., 2012). The HES consists of several integrated dimensions: social, intellectual, cultural, psychological, human, and scientific, which all have a role in it realizing its goals and objectives (Basfar et al., 2011). HES services are framed around three core elements: inputs, processes, and outputs. Inputs include the students, management, faculty members, employees, materials, infrastructures; operations include teaching, training courses, curriculum development; and outputs include quality of graduates, training programs, scientific projects, research publications, conferences, and reputation (Namor, 2012). Further, big data analytics assists universities to accurately measure their key performance indicators (KPIs) and predict their future positions, thus leading to more informed decision-making and strategy selection.
The aim of this research is to explore how huge volumes of data generated from different technologies assist HEIs with their decision-making processes. More specifically, how universities collect and analyze relevant data to ensure accurate measurements of KPIs, and thus make informed decisions. The lack of a systematic mapping study in the research arena has motivated us to conduct a mapping study. In sum, this paper makes the following contributions to the field:
- •
identifies the research frequency, progression, and type over the last decade
- •
maps the challenges of applying BDA within HES domains
- •
analyses the tools and theoretical models to support the application of BDA in the HES
These contributions provide evidence for researchers to find solutions for specific issues in applying BDA in HEIs. They also support HEIs and BDA providers to better understand the existing models, frameworks, and tools, as well as the existing gaps and challenges.
This paper is organized as follows: section 2 discusses the background and similar studies; section 3 highlights the implemented research method; section 4 presents the results of the mapping study; and section 5 concludes the paper, including suggestions for future research.