Adversarial Point-of-Interest Recommendation
Adversarial Point-of-Interest Recommendation
复制标题
DOI：
10.1145/3308558.3313609
复制
发表时间：
2019-05
期刊：
影响因子：
--
通讯作者：
Fan Zhou;Ruiyang Yin;Kunpeng Zhang;Goce Trajcevski;Ting Zhong;Jin Wu
中科院分区：
文献类型：
--
作者：
Fan Zhou;Ruiyang Yin;Kunpeng Zhang;Goce Trajcevski;Ting Zhong;Jin Wu
兴趣点（POI）推荐对于用户和企业的各种服务至关重要。人们已经开发了大量的模型来通过利用 POI 之间的各种特征和关系（例如时空、社交等）来提高推荐性能。然而，很少有研究仔细研究解释为什么用户更喜欢某些 POI 而不是其他 POI 的根本机制。在这项工作中，我们首次尝试通过提出对抗性 POI 推荐（APOIR）模型来学习用户潜在偏好的分布，该模型由两个主要部分组成：（1）推荐器（R），通过最大化这些 POI 被预测为未访问和潜在感兴趣的概率，根据学习到的分布来建议 POI； (2) 鉴别器 (D)，它将推荐的 POI 与真实的签到区分开来，并提供梯度作为在奖励框架中改进 R 的指导。通过玩极小极大游戏来共同训练两个组件，以提高自身水平，同时将另一个组件推向极限。通过进一步将 POI 之间的地理和社会关系整合到奖励函数中，并以强化学习的方式优化 R，APOIR 与最先进的方法相比，在四个标准指标上获得了显着的性能提升。
Point-of-interest (POI) recommendation is essential to a variety of services for both users and business. An extensive number of models have been developed to improve the recommendation performance by exploiting various characteristics and relations among POIs (e.g., spatio-temporal, social, etc.). However, very few studies closely look into the underlying mechanism accounting for why users prefer certain POIs to others. In this work, we initiate the first attempt to learn the distribution of user latent preference by proposing an Adversarial POI Recommendation (APOIR) model, consisting of two major components: (1) the recommender (R) which suggests POIs based on the learned distribution by maximizing the probabilities that these POIs are predicted as unvisited and potentially interested; and (2) the discriminator (D) which distinguishes the recommended POIs from the true check-ins and provides gradients as the guidance to improve R in a rewarding framework. Two components are co-trained by playing a minimax game towards improving itself while pushing the other to the boundary. By further integrating geographical and social relations among POIs into the reward function as well as optimizing R in a reinforcement learning manner, APOIR obtains significant performance improvement in four standard metrics compared to the state of the art methods.