Organisational self-identity policy analyser: An innovative AI-driven approach

Organisational self-identification (OSI) refers to a shared understanding of an organisation’s collective identity. This identity shapes the organisational culture, employee engagement and commitment, and overall performance. While policy documents define rules and interactions, both their substance and tone reinforce or diminish the OSI. However, traditional policy reviews often overlook whether the language and tone of these documents truly reflect organisational values. In this study, we introduce an AI-powered sentiment analysis framework that offers a novel and systematic approach to evaluate OSI alignment in policy texts. Our OSI Sentiment Analyser combines a customised lexicon with rule-based scoring to identify and classify sentiments at multiple levels while ensuring transparency, explainability, and respect for privacy. The analyser was validated across 78 public policies (where it achieved over 90 % agreement with expert judgement) and further tested in two NHS hospital trusts as case studies. The proposed framework not only supports evidence-based policy refinement but also fosters inclusive and value-driven communication. Its modular design promises broad application, from employee surveys to strategic organisational messaging. Accordingly, it advances both theory and practice in organisational identity analytics.