How can we use machine learning for characterizing organizational identification - a study using clustering with Picture fuzzy datasets

Abstract

This work introduces a data-driven approach based on k-means clustering with datasets elicited under a Picture fuzzy set (PFS) environment. With the vision, mission, and goals statement as a proxy for organizational identity, an actual case study is reported to demonstrate the application of the proposed approach. Four attributes were introduced to describe organizational identification: knowledge, perception, linking, and future attributes. Results revealed that out of 1,911 members, 56.67% of them are considered “proactive citizens”, 32.50% are “ambivalent citizens”, and 10.83% are “disengaged citizens”. Sensitivity analysis shows that the proposed approach is robust to changes in the model parameters. Characteristics of the types of citizens were discussed, and some managerial insights were outlined. Succinctly, this work contributes to the literature by exploring the integration of PFS to k-means clustering to characterize the extent of organizational identification of organizational stakeholders when presented with a new organizational identity – a relatively novel application in organization science.

Keywords

• Big data analytics
• k-means clustering
• Picture fuzzy sets
• Organizational identity
• Organizational identification