Article Preview
Top1. Introduction
Big data refers to the delivery of real-time insights for decision making by analysing huge amount of structured, semi-structured and unstructured data (Chen and Zhang, 2014). It has five essential characteristics: volume, velocity, variety, veracity and value (Gandomi and Haider, 2015). Since big data can help organizations to continuously improve their strategic agility while reducing the time and complexity of business operations to stay competitive in today’s rapidly-changing business environments (Demirkan and Delen, 2013). Therefore, big data has been touted as one of the most promising IT advancements by practitioners and academicians that could fundamentally change firm managers’ decision making capabilities (Nudurupati et al., 2016). The rapid development of big data markets has attracted much attention from information technology (IT) practitioners and academicians (Fasso et al., 2015). In recent years, several attempts have been made to summarize existing big data research, map its intellectual structure and predict its future directions. For example, Hashem (2015) reviewed research articles related to MapReduce for understanding the challenges on big data processing with MapReduce. Pospiech and Felden (2012) provide a state-of-the-art for the functional data provisioning of big data. Wienhofen et al. (2015) used a systematic mapping method to understand the characteristics and applications of big data and generated future research questions. Singh et al. (2015), used the analysis maps for understanding the total and growth output, authorship and country level collaboration and contributions, top publication sources, thematic and emerging themes in big data. Akter and Wamba (2016) analysed research progress of big data in the e-commerce field. These reviews provide useful information of current research on big data and facilitate the accumulation of big data knowledge. It indicates that a phase of critical introspection has begun in the big data field. This kind of big data introspection and self-reflection could be viewed as a sign of maturity of big data research. However, the rapid growth of big data research requires periodic review to keep researchers up to date. The existing reviews of big data literature are mainly based on subjective analysis of experienced research scholars in the big data field and the modern bibliometric methodology has not been leveraged to compensate for human subjectivity. Some degree of human subjectivity is indispensable to carry out review of literature (Wang et al., 2016). Yet, reviews of literature purely based on subjective analysis might be constrained by their authors' limited time, energy and cognitive capacity, and their interpretation of the literature is inevitably influenced by their personal perspectives (Raghuram et al., 2010). Therefore, it is possible that several important papers are omitted or misinterpreted to fit with the authors' own research interests (Wang et al., 2016).