3.1.4. Textual Analysis and Thematic Coding

All articles in the final sample are submitted to machine-learning-based textual analysis software named 'Leximancer'. Through Leximancer, key themes and sub-themes are identified for the framework development and further differentiation between industry and academic literature. Additional factors to the framework are included by carefully reading the articles in the final sample for robustness of the results from Leximancer and to provide context to the key themes identified by the software. Thematic content analysis through Leximancer is conducted via resource maps, and detailed results are explained in Section 5.2.

#### Resource Maps

A resource map provides a broad view of a large amount of literature in one single graph. The size of each concept point indicates its connectedness. As the algorithm goes through the list, it will attempt to draw words as close as possible to the centre of the visualization. These key themes and sub-themes identified via the resource maps and the manual systematic review of prior literature form the foundation for our DCAG.
