Advancements in News Recommendation Systems: The Role and Impact of Artificial Intelligence and Large Language Models

The rapid evolution of artificial intelligence (AI) and large language models (LLMs) significantly advanced the news recommendation systems (NRS). However, comprehensive analyses evaluating recent advancements and practical implications of integrating AI and LLMs into NRS remain scarce in existing literature. This study systematically examines AI and LLMs' effects and usage methods on NRS and analyzes 42 studies using the Prisma methodology It emphasizes the features of collaborative filtering (CF), content-based filtering (CB), hybrid systems, AI based systems including LLM-based models like BERT. While these technologies offer advanced semantic analysis and real-time adaptability opportunities, they are still partially affected by traditional challenges, such as cold-start and data sparseness, though less so than traditional methods. This study emphasizes innovations in AI-driven NRS, focusing on hybrid approaches, session-based and multi-interest models and efficient use of LLM. The findings provide actionable insights for researchers and practitioners seeking to optimize NRSs in an increasingly dynamic digital landscape.