arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.13700cs.IRcs.AIcs.SI
收藏

MAS4POI:用于下一个 POI 推荐的多智能体协作系统

MAS4POI: a Multi-Agents Collaboration System for Next POI Recommendation

Beijing Normal University-Hong Kong Baptist University United International College(北京师范大学-香港浸会大学联合国际学院);University of Warwick(华威大学)

Yuqian Wu, Yuhong Peng, Jiapeng Yu, Raymond S. T. Lee

AI总结:

本文提出 MAS4POI 系统,通过 DataAgent、Manager 等多智能体协作,利用 LLM 增强下一个 POI 推荐,并在真实数据集上验证了其提升推荐准确率的有效性。

摘要:

基于 LLM 的多智能体系统在管理跨领域的复杂决策任务方面具有潜在优势,但其在下一个兴趣点(POI)推荐中的应用仍研究不足。本文提出了一种新颖的 MAS4POI 系统,旨在通过多智能体交互增强下一个 POI 推荐。MAS4POI 支持将大型语言模型(LLM)专门化为不同的智能体,如 DataAgent、Manager、Analyst 和 Navigator,每个智能体共同参与生成下一个 POI 推荐的协作过程。该系统通过集成六种不同的 LLM 进行测试,并使用两个真实世界数据集进行评估,以提升真实场景下的推荐准确率。代码可在 https://github.com/yuqian2003/MAS4POI 获取。

英文摘要:

LLM-based Multi-Agent Systems have potential benefits of complex decision-making tasks management across various domains but their applications in the next Point-of-Interest (POI) recommendation remain underexplored. This paper proposes a novel MAS4POI system designed to enhance next POI recommendations through multi-agent interactions. MAS4POI supports Large Language Models (LLMs) specializing in distinct agents such as DataAgent, Manager, Analyst, and Navigator with each contributes to a collaborative process of generating the next POI recommendations.The system is examined by integrating six distinct LLMs and evaluated by two real-world datasets for recommendation accuracy improvement in real-world scenarios. Our code is available at https://github.com/yuqian2003/MAS4POI.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑