Research on the Development of a Virtual Simulation System for GIS Practical Teaching in Vocational Colleges

In response to the limitations of traditional GIS practical teaching, this work proposes a novel virtual simulation systemspecifically tailored for vocational education. To enhance spatial reasoning and interactive learning, we introduce GeoInstructNet, a neuro-symbolic instructional agent that integrates geospatial graph encoding with adaptive feedback mechanisms.GeoInstructNet models the spatial environment through a graph-structured encoder, preserving both topological relationsand geometric attributes of spatial primitives. By embedding real-time learner performance into a concept-guided feedbackloop, it generates dynamic, human-readable instructional hints and adjusts cognitive trajectories accordingly. We designTopographic Curriculum Navigation (TCN), an innovative instructional strategy that dynamically orchestrates task selectionbased on conceptual topography, learner deficiencies, and structural complexity. TCN employs a deficiency-guided tasksampling mechanism, balanced with complexity-controlled rejection sampling, ensuring learners progress through increasinglysophisticated spatial tasks without cognitive overload. By partitioning the curriculum into conceptual regions and trackinglearner traversal entropy, TCN adapts the educational path to individual needs while maintaining pedagogical coherence.Experimental results demonstrate that the proposed framework significantly outperforms baseline models across multipledatasets, highlighting substantial gains in accuracy, engagement, and skill retention. Our approach not only advances GISsimulation system design but also sets a new benchmark for intelligent, adaptive, and scalable GIS education in vocational settings.