Cloud-Honeypot and Differential Privacy Empowered Interactive Heat Protection against Public Opinion Analysis

With the rapid development of cloud social networks, the expression of opinions by individuals has witnessed significant growth. However, unregulated dissemination of public opinion can lead to severe consequences. This paper addresses the challenges associated with public opinion heat analysis, focusing on the potential misuse of user data by malicious entities. We propose a privacy protection mechanism that combines differential privacy theory and cloud-honeypot technologies, which is against scraper to ensure the legitimate use of data and safeguard user privacy. We propose a differential privacy mechanism with Laplace noise, which adds Laplace noise to the number of reposts, comments, and likes of the Weibo posts published by users, making it impossible for attackers to calculate the correct interactive heat and thus protecting the privacy of Weibo interaction popularity. At the same time, we propose a privacy protection mechanism which is applicable in combination with honeypot technology. After the cloud server identifies that the visitor is a malicious scraper, it redirects them to the honeypot webpage. This modified webpage incorporates additional Laplace noise, thereby enhancing the security measure. Additionally, we design an experiment to simulate the impact of adding noise to the data and compare the results to verify the feasibility of the differential privacy mechanism in countering scrapers. Finally, we discuss future directions for anti-scraping strategies based on differential privacy.