Content area
Full text
Introduction
Service researchers are faced with an unprecedented volume of textual data generated from a range of sources and formats such as research publications, news items, industrial reports, online chatter, surveys, interviews, blogs, scripts and notes. It is expected that the number and complexity of these qualitative data documents will only increase in the future. The International Data Group predicts that by 2025 there will be 163 ZB of data in the world, with around 80 per cent of business-relevant information originating from unstructured forms, primarily text (Techrepublic, 2017). Key sources of this growth are from firm-generated content including, for example, annual reports news items, and from customers’ comments on websites and consumer-generated content that appear in social networking sites.
Methods of conducting and analyzing qualitative research are also changing. Data has become more readily available and tools such as text mining are well suited to handle large quantities of unstructured data and extract knowledge from these disparate primary and secondary data sources in short periods (Hartmann et al., 2016; Humphreys and Rebecca, 2017; McColl-Kennedy et al., 2019; Rust and Huang, 2014; Villarroel Ordenes et al., 2014; and Zaki and Neely, 2019). Text mining uses a set of natural language processing (NLP) and machine learning techniques to process textual documents, derive patterns within a structured format and provide evaluation and interpretation of the output to gain insights that matter (Feldman and Sanger, 2006). It involves information retrieval, lexical analysis to study word frequency distributions, information extraction and machine learning techniques including visualization and predictive classification analytics (Schmunk et al., 2014). Despite their importance, these techniques, while well established in information systems and computer science literatures, are less well known in service research, and therefore many service researchers do not know how to apply the techniques. Further, there is little research to guide service scholars and practitioners on how to determine which methods are most appropriate in different contexts.
Hence, the contribution of this article is threefold:
to provide a text mining analysis roadmap (TMAR) on how to use text mining methods in practice;
to demonstrate the usefulness of text mining techniques to analyze qualitative data at the different research stages; and
to illustrate how service researchers can generate insights that result...