<p>Overlapping community detection algorithms play a significant role in social network analysis due to their ability to more accurately describe social relationships resulting from human activities. However, excessive mining of social network data inevitably leads to privacy leakage issues. Although community hiding algorithms have been proposed to address privacy concerns, existing methods are primarily focused on countering community structures detected by non-overlapping community detection algorithms, with limited effectiveness against overlapping community detection algorithms. To address this limitation, we propose a novel overlapping community hiding algorithm. First, considering the unique characteristics of overlapping nodes, we propose the SP (dispersivity) function to measure the distribution of nodes within overlapping structures. Furthermore, we introduce the OCSP (overlapping community hiding algorithm based on SP) algorithm. The OCSP algorithm first utilizes the inherent characteristics of communities to extract prior information, thereby reducing the search space of the algorithm. Then, it combines the SP method as a loss function with a simulated annealing algorithm to achieve the hiding of overlapping community structures. Finally, the transferability of the algorithm is verified by applying the link combinations obtained from a specific overlapping community detection algorithm to other community detection algorithms, resulting in good hiding effects. Experimental results on five real-world datasets and four overlapping community detection algorithms demonstrate that our hiding algorithm exhibits excellent interference efficiency against overlapping community detection.</p>