Semantic structures of business analytics research : applying text mining methods

Introduction. Business analytics has grown exponentially over the last decade, combining technologies, systems, practices and applications. It has attracted both practitioners and academics based on its capabilities to analyse critical business data to gain new insights about business operations and the market. The research goal of this paper is to identify major research topics and trends using text mining techniques.

Method. We applied text mining methods such as co-word analysis and topic modelling to 1,024 published research documents in the business analytics field found in the Web of Science and Scopus databases.

Analysis. We used term co-occurrence maps and latent Dirichlet allocation to mine and visualise data.

Results. Findings showed that the knowledge structure of business analytics consists of three main themes: analytical methods, business analytics in practice, and business analytics value creation. Big data analytics, machine learning, and data science techniques are major topics. Further, business analytics research topics were identified and clustered into four thematic groups.

Conclusions. The findings present a context for business analytics research development. They show recent trends in the field, namely a predominant interest in big data analytics, social networks, business value, the health sector, and customer retention.