Scholarlink

In today’s modern world, communications through social media have replaced conventional communication scenarios. Almost 70% of the world’s population uses social media to share life events, impart knowledge, manage invites, etc. Due to this, it is essential to validate the authenticity of social media users. Currently, around 1.5 million fake accounts are on Facebook, which causes the spreading of misinformation and fake news. Researchers have come up with a plethora of models to detect fake currency; these models differ from one another in terms of textual subtleties, functional benefits, deployment-specific constraints, and application-specific potential future reaches.Researchers have come up with a plethora of models to detect fake currency; these models differ from one another in terms of textual subtleties, functional benefits, deployment-specific constraints, and application-specific potential future reaches. With this variation comes varied output in terms of their quantitative & qualitative performance as in scalability, accuracy, response time, computational complexity etc. Because of such a wide variation in performance, social media designers are compelled to validate several models for individual applications. This increases the delay & cost needed to deploy the final model for practical scenarios. Thus, it is challenging to identify optimal fake-social-media profile detection models for performance-specific & functionality-specific deployments. To overcome these issues and simplify the task of fake social-media-profile model selection, this text initially discusses existing models in-depth based on their functionality. After this discussion, an empirical & pragmatic analysis of these models is presented. Researchers are evaluated in terms of the accuracy of fake profile detection, delay needed for research, computational complexity, scalability, and cost of deployment. Based on this comparison, researchers can identify optimal fake-profile detection models. This text also proposes calculating a novel Fake Profile Detection Rank Metric (FPDRM), which combines these parameters to identify models that can be used for higher overall performance levels under different use cases.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Pragmatic Review and Implementation of Models Targeted Toward the Identification of Fake Social Media Profiles

  • 章节

  • 发表日期：

    2025年

查看全文
  • Bhrugumalla L. V. S. Aditya,
  • Sachi Nandan Mohanty

摘要

In today’s modern world, communications through social media have replaced conventional communication scenarios. Almost 70% of the world’s population uses social media to share life events, impart knowledge, manage invites, etc. Due to this, it is essential to validate the authenticity of social media users. Currently, around 1.5 million fake accounts are on Facebook, which causes the spreading of misinformation and fake news. Researchers have come up with a plethora of models to detect fake currency; these models differ from one another in terms of textual subtleties, functional benefits, deployment-specific constraints, and application-specific potential future reaches.Researchers have come up with a plethora of models to detect fake currency; these models differ from one another in terms of textual subtleties, functional benefits, deployment-specific constraints, and application-specific potential future reaches. With this variation comes varied output in terms of their quantitative & qualitative performance as in scalability, accuracy, response time, computational complexity etc. Because of such a wide variation in performance, social media designers are compelled to validate several models for individual applications. This increases the delay & cost needed to deploy the final model for practical scenarios. Thus, it is challenging to identify optimal fake-social-media profile detection models for performance-specific & functionality-specific deployments. To overcome these issues and simplify the task of fake social-media-profile model selection, this text initially discusses existing models in-depth based on their functionality. After this discussion, an empirical & pragmatic analysis of these models is presented. Researchers are evaluated in terms of the accuracy of fake profile detection, delay needed for research, computational complexity, scalability, and cost of deployment. Based on this comparison, researchers can identify optimal fake-profile detection models. This text also proposes calculating a novel Fake Profile Detection Rank Metric (FPDRM), which combines these parameters to identify models that can be used for higher overall performance levels under different use cases.