抄録
With the accelerated advancement of digitalization and platformization, the spatial representation of human emotions is shifting from predominantly experiential and qualitative paradigms toward data-driven, computable, and visual analytical frameworks centered on social media digital footprints, thereby driving profound transformations in both the theoretical construction and methodological systems of emotional geography. From a review perspective, this article systematically synthesizes recent progress in social media–driven emotional geography, focusing on the evolution of theoretical foundations and key concepts, types of social media data and emotion recognition approaches, multiscale emotion–space coupling modeling, as well as the representation and application of emotional maps and emotional landscapes. Existing studies have achieved important breakthroughs in revealing the spatiotemporal heterogeneity, dynamic evolution, and social embeddedness of individual and collective emotions; however, significant limitations remain in terms of data representativeness, cross-platform comparability, methodological robustness, theoretical interpretability, and ethical and privacy governance. The applicability boundaries, complementary relationships, and synergistic potential among different technical pathways and research paradigms have yet to be systematically compared and integrated within a unified framework. Accordingly, this review further summarizes the common challenges and key controversies in current research and outlines future directions from the perspectives of multisource data integration, methodological synergy and optimization, theoretical deepening, and responsible research practices. This article aims to provide a systematic, critical, and forward-looking review framework for emotional geography in the digital era, offering a reference for theoretical innovation, methodological integration, and practical application in related fields.
