Maximum Recommendation in Geo-social Network for Business

• Main Authors: Zongmin Cui, Jing Yu, Sanggyun Na
• Format: Article
• Language: English
• Published: Faculty of Mechanical Engineering in Slavonski Brod, Faculty of Electrical Engineering in Osijek, Faculty of Civil Engineering in Osijek 2019-01-01
• Series: Tehnički Vjesnik
• Subjects: business policy; geo-social network; maximum recommendation; negative influence
• Online Access: https://hrcak.srce.hr/file/320433

Most of existing methods do not consider the maximum recommendation issue. Meanwhile, the methods also do not consider the negative influence in recommendation model. These two shortcomings limit further application of the recommendation system. In another word, the shortcomings not only decrease the recommendation effect but also increase the recommendation cost in the business. To remove the shortcomings, we propose a Maximum Recommendation scheme in Geo-social network for business (called as MRG). On the one hand, we identify k nodes with maximum recommendation according to the expected paid node number k. On the other hand, we exclude the negative node from the geo-social network. Based on the above innovation, we effectively increase the recommendation effect and decrease the company's recommendation cost. Meanwhile, MRG considers the negative influence to enhance the recommendation efficiency. Experimental results show that our scheme has better performance than most of the existing methods in the maximum recommendation field.