Scholarlink

The integrity and security of online platforms depend heavily on the ability to identify false profiles. Several studies have been conducted to identify phony profiles on various social media platforms. But, each of the works has some limitations such as dataset and accuracy. In order to solve the problem of detecting fake profiles, we offer a novel ensemble model in this research that combines the Random Forest (RF) algorithm, with a bagging classifier, and the Long Short-Term Memory (LSTM) deep learning (DL) model. Our model’s remarkable accuracy rates for the Machine Learning (ML) and LSTM components are 94% and 92%, respectively. To fully evaluate the model’s performance, we also analyze the precision, recall, and F-score. A promising method for improving false profile identification on online platforms is the proposed ensemble model. Its dependability and efficiency are demonstrated by its accuracy, precision, recall, and F-score measurements. The results of this study can help social media platforms and online communities protect the security and trust of their users by improving profile authenticity verification. We have done our research on Instagram fake profile detection.

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

Fake Profile Detection of Social Media Using Deep Learning and Ensemble Machine Learning

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • U. Sivaji,
  • B. Rupa Devi,
  • L. N. C. Prakash K,
  • K. Reddy Madhavi,
  • M. Reddi Durgasree,
  • Shaik Mohammad Rafee

摘要

The integrity and security of online platforms depend heavily on the ability to identify false profiles. Several studies have been conducted to identify phony profiles on various social media platforms. But, each of the works has some limitations such as dataset and accuracy. In order to solve the problem of detecting fake profiles, we offer a novel ensemble model in this research that combines the Random Forest (RF) algorithm, with a bagging classifier, and the Long Short-Term Memory (LSTM) deep learning (DL) model. Our model’s remarkable accuracy rates for the Machine Learning (ML) and LSTM components are 94% and 92%, respectively. To fully evaluate the model’s performance, we also analyze the precision, recall, and F-score. A promising method for improving false profile identification on online platforms is the proposed ensemble model. Its dependability and efficiency are demonstrated by its accuracy, precision, recall, and F-score measurements. The results of this study can help social media platforms and online communities protect the security and trust of their users by improving profile authenticity verification. We have done our research on Instagram fake profile detection.