Enhancing the human sensemaking process with the use of social network analysis and machine learning techniques

Abstract

Sensemaking is often associated with processing large or complex amount of data obtained from diverse and distributed sources. Sensemaking enables leaders to have a better grasp of what the data represents and what insights they can get from it. Thus, sensemaking is considered extremely important in mature markets where the competition is fierce. To-date, the research base on sensemaking has not moved far from the conceptual realm, however. In response, this research provides a conceptual framework that explains the core processes of sensemaking – noticing, interpretation and action – and examines how emerging technologies such as Social Network Analysis (SNA) and Machine Learning (ML) techniques help to enhance the human sensemaking process in generating valuable insights during data analysis. Design Science Research (DSR) is adopted as a research methodology in the context of financial transactional data analysis, aiming to make sense of the data while exploring conceptions of customer value for a mainstream commercial bank alongside the perceived need for banking products. Three analytical models are introduced, examining Connected Customer Lifetime Value (CCLV), Network Relationship Equity (NRE) and product purchasing frequency based on customer ‘personas’. The former models employ SNA techniques in providing novelty, the latter combines the outcomes of SNA with ML clustering algorithms to provide a base on which product holdings and purchase frequency analysis are overlaid – providing a novel form of recommendation. Ongoing evaluation of the developed models is used to explore the nuances of the sensemaking process and the ability of such models to support that process (in the given domain).