With the rapid development of artificial intelligence technology, its application in the maritime domain is becoming more and more common. Effectiveness assessment of maritime cluster cooperative tasks is an indispensable part of validating maritime mission plans. Traditional assessment methods rely on experts' experience, which is not only time-consuming but also influenced by subjective factors, so more scientific and efficient assessment methods are needed. In this paper, a mesh structure index system for effectiveness assessment based on OODA decision chain is proposed, and a residual self-attention network is used to construct an algorithmic model for effectiveness assessment. The feasibility and effectiveness of the method are verified through simulation experiments, aiming at objectively and effectively assessing the effectiveness of complex systems, while reducing the dimensional catastrophe caused by multi-dimensional data input and the subjective influence in the process of assessing indicator feature extraction. The results of the study show that the method can significantly improve the efficiency and accuracy of the effectiveness assessment process for maritime cluster cooperative tasks.

effectiveness assessment,mesh structure index system,maritime cluster cooperative tasks,deep learning,supervise contrastive learning