AI-Driven Decision Support for Global Knowledge-Sharing Interventions

Abstract

In global organizations, knowledge-sharing interventions — such as webinars, workshops, and targeted training sessions — are essential drivers of collaboration, skill development, and organizational learning. However, decisions about which type of intervention to conduct are typically made through intuition rather than systematic, data-driven analysis. This paper proposes an AI-driven Decision Support System (DSS) that analyzes organizational signals — including stakeholder feedback, attendance history, engagement metrics, survey responses, cultural factors, and Natural Language Processing (NLP)-derived behavioral patterns from meeting transcripts — to recommend the most suitable knowledge-sharing intervention. The system classifies meeting discussions into behavioral pattern categories such as Conceptual Knowledge Gap, Technical Failure, Governance Issue, and Exploration Mode, and maps each pattern to a targeted intervention strategy. A case study drawn from real meeting transcripts within a Microsoft Power Platform deployment context validates the model, demonstrating high classification confidence for conceptual gap identification. The proposed framework offers a replicable, data-informed approach that can improve decision quality, reduce ineffective sessions, and optimize organizational learning outcomes across culturally diverse global teams.

Keywords