This study addresses the demand for socialized sharing of ice and snow sports resources in higher education institutions in Heilongjiang Province, proposing a “dual-core driven” development model. It constructs an intelligent resource integration platform based on the k-nearest neighbor clustering algorithm and optimizes teacher participation mechanisms using an agency incentive model. Through simulation experiments, the method is validated to significantly enhance resource sharing efficiency. The k-nearest neighbor clustering algorithm achieves 99.74% accuracy and 98.98% recall rate (converging after 30 iterations), representing a 6-percentage-point improvement over traditional methods. Under 500 concurrent users, resource sharing speed reaches 583.94 MB/s, with a peak throughput of 760.05 MB/s, marking a 31.7% improvement over cloud computing solutions. The sports resource information integration system constructed in this paper demonstrates strong transmission stability, with a resource loss rate of only 3.6% under high-concurrency scenarios (1,000 users), which is 1.4-2.6 percentage points lower than the comparison method. Case analysis shows that resource integration efficiency reaches 99.17%, with an average efficiency exceeding 98% across the five major resource categories. Empirical evidence indicates that this model effectively addresses the challenges of scattered, low-quality, and stagnant winter sports resources through technological innovation and mechanism coordination, providing a scalable path for the socialized development of winter sports in cold-region universities.
