부산대학교 도서관

A personalized recommender system meticulously analyzes users’ data encompassing their purchases, ratings, and interactions with other users. This thorough analysis allows the system to provide personalized and tailor-made recommendations aligned with users’ preferences. The advent of IoT-connected sensors has significantly augmented the availability of contextual information, prompting the rapid development of Context-Aware Recommendation Systems. Consequently, we have embraced a cloud computing approach, specifically leveraging fog architecture, which optimizes shared resources and offers a stable environment for consistent performance measurement. The primary objective of this paper is to propose an intelligent context-aware recommender system based on fog architecture, employing hybrid machine learning approaches. This innovative system involves gathering data from fog nodes and employing a process that learns users’ profiles by screening and analyzing reflective feedback loops. Ultimately, this approach enables the prediction of personalized items.