On Selecting Vehicles as Recommenders for Vehicular Social Networks

Vehicular social networks, as one type of new future networks, integrate mobile wireless communications and social networks together, and they have attracted a plenty of interest from researchers. This paper argues that vehicular social networks joined with cloud and fog computing platforms can serve as recommenders and play a significant role in social lives. A new application for recommender systems that selects vehicles as recommenders is proposed to support marketers in improving marketing effectiveness. Then, we propose three algorithms to enhance the marketers' marketing effectiveness based on different evaluation standards. The first algorithm selects those vehicles that can obtain maximum benefits for the marketers, in which the vehicles are selected based on passing regions with more benefits. However, this selection method may lose some potential markets because of region limitations. The second algorithm selects those vehicles that can reach the maximum coverage ratio in the city and bring the marketers more marketing effectiveness in the future, although the current benefits are not the best. The third algorithm combines the two prior algorithms, finding a tradeoff between coverage and benefits. We finally evaluate the effectiveness of the proposed algorithm with real-world data to show the effectiveness and efficiency of the "On Selecting Vehicles as Recommenders for Vehicular Social Networks" scheme: for the scheme based on the benefit factor, the performance criteria of the benefit ratio can be improved by 21% over the existing selection methods; for the scheme based on the coverage factor, the performance criteria of the coverage ratio can be improved by 13% over the existing selection methods; and for the third selection algorithm based on the two factors, the performance criteria of the benefit ratio and the metric 9 can be improved by 18.9% and 13.7%, respectively.