城市多源数据投放位置推荐算法的研究

Research on Urban Multi-source Data Placement Recommendation Algorithm

本文融合城市出租车交通数据、城市路网数据以及城市兴趣点(Point of Interest,POI)数据,通过刻画用户出行的潜在意图,提出了面向潜在目标用户的户外广告影响力量化模型;同时基于城市商业基础设施与公共交通设施分布情况,设计了户外广告在不同空间区域投放的定价策略。在此基础之上,形式化定义了广告影响力-投放成本两优化目标的户外广告精准投放位置推荐问题,通过对传统GSkyline查询算法的改进,基于最大支配数计算以及高效剪枝策略设计了位置推荐查询算法。在真实数据集上的实验结果验证了本文所提相关方法的有效性。

This paper combines urban taxi traffic data,urban road network data and city point of interest(POI)data.By portraying the potential intention of user travel,this paper proposes an outdoor advertising influence power model for potential target users.At the same time,based on the distribution of urban commercial infrastructure and public transportation facilities,the pricing strategy of outdoor advertising in different spatial regions was designed.On this basis,this paper formalizes the recommendation of outdoor advertising precision placement for advertising influence-delivery cost optimization goal.By improving the traditional G-Skyline query algorithm,the location recommendation query algorithm is designed based on the maximum dominance calculation and efficient pruning strategy.