Secure Information Foraging Using Fully Homomorphic Encryption and AGNES Clustering

In the ever-expanding landscape of social media, users struggle with navigating an overwhelming volume of information. This research introduces an innovative approach to Information Foraging by incorporating data encryption as a core component-a novel perspective never before explored in this context. The goal is to fortify the confidentiality and integrity of users’ critical data, setting a standard for safeguarding information from external threats. Within this work, we employ Fully Homomorphic Encryption in conjunction with AGNES clustering. While Fully Homomorphic Encryption ensures robust data protection, the model’s efficiency is guaranteed through a hierarchical clustering structure facilitated by AGNES. The evaluation was carried out on a dataset encompassing over 900,000 posts obtained from the social network X, covering a diverse array of topics. The results underscore the model’s competence in efficiently and securely identifying relevant information while upholding users’ privacy. Furthermore, a comparative analysis with existing approaches from the literature highlights the superiority of our proposal, establishing a new frontier in the integration of data encryption within the Information Foraging paradigm.