A Link Prediction Method Based on Neural Networks

Link prediction to optimize network performance is of great significance in evolution. Because the complexity systems and uncertainty evolution, it faces many challenges. This paper proposes a new link method based on neural networks trained scale-free as input data, optimized by models output data. In order solve influence generalization experiments, greedy pruning strategy applied. We consider efficiency proposed global structure reliability objectives comprehensively evaluate advantages method. The experimental results demonstrate that generates with better than traditional models.