CONFERENCE / ICICMCT'23

For industrial production in the Internet of Things, a mobile industrial network is essential. It ensures that equipment will operate normally and that industrial production will normalize. However, spammers may take use of this trait to harm people and impede business operations. Spammers are users who solely spread spams, such as links to viruses and adverts. Spammers have formed organizations in order to maximize their benefits as mobile network membership has increased, which has led to confusion and significant losses in industrial productivity. Due to the properties of multidimensional data, it is challenging to separate spammers from regular users when using a dataset from a cloud server. According to the simulation findings, SIGMM performs better in terms of recall, accuracy, and temporal complexity than these earlier systems. This research offers a Spammer Identification strategy based on Gaussian Mixture Model (SIGMM) for industrial mobile networks in order to solve this issue. It delivers accurate spammer identification without depending on fluid and erratic interactions. SIGMM integrates the data presentation with the model generation process, which assigns a class to each user node. Using a mobile network dataset from a cloud server, we compare SIGMM with the reality mining method and hybrid FCM clustering technique to validate it. According to the simulation findings, SIGMM performs better in terms of recall, accuracy, and temporal complexity than these earlier systems.