Data-Driven Policy Making Framework Utilizing TOWS Analysis

In the era of hyper-uncertainty, the demand for data-driven policy making has grown significantly, as it enables the fast and accurate capture of complex phenomena. This contrasts with conventional policy making processes, which are often slow and heavily reliant on the limited experience and knowledge of a select group of experts. Although interest in data-driven approaches continues to grow, existing research primarily focuses on identifying policy agendas, leaving policymakers to develop the specific content for those agendas from scratch. In this paper, we propose a novel data-driven policy making framework based on TOWS analysis, a well-established method for strategic extraction that systematically integrates internal strengths and weaknesses with external opportunities and threats. Our framework automatically extracts policy agendas and their specific contents. It begins by identifying national capabilities and cross-cutting issues to establish key agendas using topic modeling and sentiment analysis. Then, a TOWS matrix is constructed to derive policy directions by linking capabilities with issues. Here, relevant phrases are extracted to elaborate on these directions, forming policy content. To validate our framework, we conducted a case study on South Korea’s post-COVID-19 policies. The policies derived from our framework closely aligned with real-world ones and, in some instances, provided more detailed recommendations. This demonstrates the potential of data-driven approaches to enhance the precision and comprehensiveness of policy development in times of uncertainty.