Bridging Perceived and Projected Destination Images: A Machine Learning-Driven Systemic Network Analysis

Abstract

Understanding the differences between the projected and perceived images of destinations plays a critical role in destination marketing and the fulfillment of tourists' expectations. However, existing research still lacks a systematic explanation of the network-level differences in destination image, and tends to overlook the underlying influence mechanisms among its factors. Using Macau as a case study, this research develops a decision-making model to deepen the understanding of interrelationships among destination image factors, and to explore the structural network differences between images. To achieve this, the research employs Latent Dirichlet Allocation (LDA), quadrant analysis, and the Decision-Making Trial and Evaluation Laboratory (DEMATEL) method. This study offers a systematic perspective to narrow the gap between the projected and perceived image. The results enable DMOs to gain deeper insights into tourists’ needs and perceptions, as well as to identify the key factors shaping destination image, and the internal influence relationships among its components.