Improving the Security and Confidentiality in the Internet of Medical Things Based on Edge Computing Using Clustering

Table 1

An overview of previous models and the advantages and disadvantages of each.

Disadvantages	Advantages	Model (method)
Lack of privacy in some cases.	More accuracy and expansion of information in less time.	A model based on fuzzy logic, artificial intelligence, and gray wolf [9].
The process of searching and exchanging data is associated with problems in relatively large volumes of data.	More precision in the cluster social network.	Clustering model of social networks using multilayer perceptron network and gray wolf [10].
Dissemination of confidential information that owners are reluctant to disclose.	Wider use of social media, viral marketing, in which various forms of marketing message related to a company, brand, or product are transmitted with exponential growth, often through the use of social media applications.	A hybrid model of standard algorithm (naive k-means) and bee and gray wolf algorithm [11].
Time-consuming and repetitive results in the third stage and later, which is associated with high error.	Drawing a decision tree from data, expand the dissemination of information in social networking associations, break down a resource set into subcategories based on testing the value of an attribute.	Decision tree and gray wolf model [13].
There is a mapping of all related edges among the studied heads that the genetic algorithm is unable to examine.	Utilizing the capacity of social networks at many individual and social levels to identify problems and determine their solutions, establishing social relations, managing organizational affairs, policy-making, and guiding individuals on the path to achieving goals.	Model based on genetic algorithm and gray wolf [14].
Increased computational complexity compared to previous methods.	High accuracy and efficiency of the algorithm.	A model based on new genetic algorithm and gray wolf [15].
Occurrence of shilling attacks and loss of users’ trust in recommender systems.	More speed in social networking forums.	Model based on participatory refinement algorithm and gray wolf [12].
Precise settings of input parameters that may be associated with errors.	Improving the quality of community discovery.	Model based on F algorithm and gray wolf [16].
Identifying edges statically, disregarding user interactions.	Maintaining data privacy.	Model based on a cuckoo optimization algorithm and gray wolf algorithm [17].
When the number of users is large, the number of errors in the algorithm exceeds the allowable limit defined in the default algorithm, and the results become invalid.	Improving the structure of users’ social network, providing different services to users spatially.	Model based on a new algorithm based on spatial data and gray wolf [18].
Loss of time and place and more role of space in the social network and shaking the foundations of users’ identities.	The social networks studied in the communication process can have positive effects.	Model based on genetic algorithm with a generational division and gray wolf [19].
The complexity of the implementation process and its time-consumption, the occurrence of human error in computing.	High accuracy and speed in social network processing.	A hybrid model of clustering and artificial neural network and gray wolf [20].
Lack of coverage of all nodes in the forum on large social networks.	Identifying the intermediate centrality of communication channels with anonymity and the characteristics of each anonymous node and effective communication channels, maintaining data privacy.	Combined model of anonymity clustering and genetic algorithm and gray wolf [21].
Inability to check all nodes and community members on social networks.	Identifying how information is disseminated on social networks, identifying key nodes in this type of network, providing quality control of information dissemination on social networks.	Model based on the evolutionary algorithm of bees and gray wolf [22].
Central network nodes cannot be checked by this method.	Identification and mapping of information exchange networks in social networking associations, revealing the main nodes.	Model based on modified genetic algorithm and gray wolf [23].