Technical gazette, Vol. 26 No. 2, 2019.
Original scientific paper
https://doi.org/10.17559/TV-20181105122320
Maximum Recommendation in Geo-social Network for Business
Jing Yu
; School of Information Science and Technology, Jiujiang University, No. 551, Qianjin East Road, Jiujiang, Jiangxi 332005, China / College of Business Administration, Wonkwang University, No. 460, Iksandae-ro, Iksan, Jeonbuk 54538, Korea
Sanggyun Na
; College of Business Administration, Wonkwang University, No. 460, Iksandae-ro, Iksan, Jeonbuk 54538, Korea
Zongmin Cui
; School of Information Science and Technology, Jiujiang University, No. 551, Qianjin East Road, Jiujiang, Jiangxi 332005, China
Abstract
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.
Keywords
business policy; geo-social network; maximum recommendation; negative influence
Hrčak ID:
219534
URI
Publication date:
24.4.2019.
Visits: 1.736 *