Hybrid Rural Landscape Characterization and Typological Governance Strategies in Metropolitan Fringe Areas Based on Machine Learning: A Case Study of Baoshan District, Shanghai

Rapid urbanization and industrialization have significantly reshaped rural landscapes in metropolitan fringe areas, resulting in “hybridized” characteristics. This study establishes an analytical framework to systematically characterize hybrid rural landscapes, diagnose specific local issues, reveal their spatial differentiation patterns and driving mechanisms, and propose targeted governance strategies. Taking 124 rural units in Baoshan District, Shanghai as a case, multi-source data from the latest available years (2020–2023) were compiled as a cross-sectional snapshot, and a comprehensive indicator system integrating landscape pattern (P), social function (F), and spatial vitality (V) was developed. Utilizing multi-source geospatial data—including land-use maps, points of interest, and mobile signaling data—Gaussian Mixture Models were applied to classify typical hybrid landscape types. Spatial evolution processes and underlying driving forces were further interpreted through remote sensing imagery analysis, field investigations, and policy document reviews. Eleven distinctive hybrid rural landscape types (HTs) were characterized, forming a spatial gradient from urban to rural, encompassing “high-density urbanized” → “ecologically embedded” → “production–living integrated” → “traditional rural landscapes”. Additionally, five representative evolutionary patterns—“urban restructuring”, “ecological orientation”, “industrial-driven transition”, “transitional hybridization”, and “traditional preservation”—were identified, shaped by spatial configuration, planning policies, industrial investments, and demographic dynamics. The framework enhances understanding of the complexity and evolutionary dynamics of rural landscapes, providing theoretical insights and practical guidance for effective typological governance and targeted policy interventions.