ABSTRACT

In recent times, location recommendation has received significant attention from the researchers due to emerging utilisation of Location Based Social Networks in the prediction process. In this paper, we present a new multi-agent based framework to generate better-personalised location recommendations. We address the personalisation problem through the dynamic user profile that incorporates the user’s long-term and short-term cognitive behaviour. The better adaptation of user cognitive behaviour enhances the prediction process and improves overall user experience with better recommendations. A detailed user study is conducted to reveal the improved performance of proposed approach through enhanced recommendations in comparison with other approaches.

摘要

近年来，由于在预测过程中新出现的基于位置的社交网络的使用，位置推荐受到了研究人员的极大关注。在本文中，我们提出了一种新的基于多代理的框架来生成更好的个性化位置推荐。我们通过包含用户长期和短期认知行为的动态用户配置文件来解决个性化问题。用户认知行为的更好适应增强了预测过程，并通过更好的推荐改善了整体用户体验。进行了详细的用户研究，以通过与其他方法相比增强推荐来揭示所提出方法的改进性能。