An Integrative Computational Intelligence for Robust Anomaly Detection in Social Networks

Anomaly detection is one of the most important tasks for maintaining the integrity, security, and trustworthiness of online communities in a social network. This paper proposes AdaptoDetect, which represents a new framework; it discusses a new anomaly detection approach called Pufferfish Optimization Technique for feature selection, together with a Graph Embedding Autoencoder for identifying anomalies. What makes AdaptoDetect special is that, with the use of POT, it has a distinctive capability in dynamic adaptation against network changes by selecting only the most relevant features in social network data. The technique for optimization underlines the important attributes for anomaly detection so as to allow a more fine-tuned and accurate identification process. Meanwhile, GEAE effectively learns low-dimensional representation of graph nodes, capturing complex patterns and interrelations in the structure of graphs. These graph embeddings further enhance anomaly detection by highlighting deviation from standard social network behaviors, hence making the detection of those irregularities more accurate. The novelty in this integration of POT and GEAE makes AdaptoDetect a strong, adaptive framework suited for tackling the dynamic nature of social networks. Extensive evaluations over various social network datasets and scenarios show the superior performance of AdaptoDetect compared to state-of-the-art methods, especially regarding its adaptiveness to the alteration in networks and detection of anomalies with high accuracy. Besides fortifying the security of social networks, making online environments much safer and more trustworthy will be contributed to by significantly enhancing resilience and reliability in social networks. © 2024 College of Education, Al-Iraqia University. All rights reserved.