Recently, detecting dynamic patterns for growing communities in social networks have attracted significant attention. The objective of dynamic community detection is to analyze and identify clusters in complex networks that change over time. Different optimization algorithms, including both single-objective and multi-objective approaches, have been employed to address the challenge of dynamic community detection. Although current evolutionary algorithms for identifying community structure can traverse extensive areas of partition space, they often become stuck in local minima. In addition, limited studies have addressed this issue by integrating local search strategies with evolutionary algorithms for community identification. This paper introduces an enhanced mutation strategy, known as vulnerable node reassignment (VNR), in order to improve the search efficiency of the Multi-Objective Evolutionary Algorithm with Decomposition (MOEA/D) and speed up the convergence of the algorithm. This algorithm was tested on two synthetic and two real life datasets, under three measures NMI, Conductance and modularity (Q). Our heuristic was applied to the Kim and Han dataset, and the Q model captured community structure effectively, NMI averages 0.99 in Fix-Net z3 and 0.98 in Var-Net. Furthermore, Q shows the high reduction in Conductance values between before and after heuristic, reaching 0.41 on Fix-Net z3. On MS-Net and EC-Net, the expansion (EX) averaged 0.95 and 0.82 NMI. The Q model performs well using the Q measure, averaging 0.60 in the Cell phone and 0.48 in the Enron mail. Additionally, ANOVA, Tukey tests and Standard deviation applied on all datasets. The results demonstrated that our enhanced algorithm was outperformed several states- of- the-art algorithms.
