Exploration and Reflection on Precise Recommendation in Complex Information Environments: Boundaries, Challenges, and Future Prospects

As a proactive filtering technique, recommender systems (RSs) analyze users’ prior preferences to assist them in selecting valuable information from a vast amount of data. It is a crucial tool for handling information and cognitive overload. However, with the development of digital and communication technologies and shifts in user awareness, existing RSs face new limitations and challenges, including data violations, algorithmic discrimination, information cocoons, etc., making them inadequate to meet the diverse application requirements and user information needs across various scenarios. Therefore, a more effective RS is required. This study aims to identify several present limitations and challenges faced by RSs and provide insights for the design of future RSs. Firstly, some basic concepts and techniques of RSs are introduced. Then, several significant limitations and challenges in designing an effective RS are investigated and reported in terms of data, algorithms, and evaluations. Finally, opportunities for further study are proposed in response to the problems reported.