This paper proposes a basic education group school collaborative management decision support system for high-quality and balanced development. The system integrates multi-source governance data modeling, index fusion, intelligent evaluation and recommendation feedback, and supports resource allocation, inter-school collaboration, teacher support and supervision. A data set covering three junior middle schools, five primary schools and four school districts was constructed, which contained 1236 teachers, 14508 students, 326 courses and 18460 governance events. A calculation process combining graph structure coding, weighted aggregation and feedback correction was formed. Experimental results show that the system achieves 93.4% accuracy of decision recommendation, 91.2% F1 value of inter-school balance recognition and 0.887 consistency of collaborative state on cross-school governance tasks, and reduces the average response delay from 11.6 s to 2.9 s under 300 concurrent loads. The accuracy of resource matching is improved by 16.8%, and the satisfaction of management subjects is improved from 7.1 to 8.9, which provides a computable and operational framework for data-driven governance of the group school.
