多粒度犹豫模糊语言信息下的群推荐方法

Multi-granular hesitant fuzzy linguistic term sets and their application in group recommendation

针对群推荐系统的偏好信息来自不同的网络平台,被推荐项目具有多粒度性、犹豫模糊性和多属性等问题.本文首先定义了多粒度犹豫模糊语言术语集的概念,定义了多粒度犹豫模糊语言术语集的广义距离公式、广义豪斯多夫距离公式和广义混合距离公式;其次,考虑被推荐项目属性的权值问题,定义了相应的广义加权距离公式距离、广义加权豪斯多夫距离公式和广义混合加权距离公式等.研究了这些公式的性质,讨论了公式之间的关系;最后,将这些距离公式结合满意度公式用于群体推荐问题,并进一步分析了公式的参数对满意度及电影排序的影响情况.

Concerning to the problem that the preference information of the individuals from different groups coming from different network platforms, and the individuals of different groups are often hesi- tant and fuzzy when they express the preference information of the multiple criteria recommended items. Fristly, the concept of multi-granular hesitant fuzzy linguistic term sets （MHFLTS） is defined, and different types of distance measures of between two MHFLTSs are proposed, which based on the traditional distance measures such as the Hamming distance, the Euclidean distance, and the Hausdorff metric. Secondly, a variety of weighted distance measures between two collections of MHFLTSs are proposed and the prop- erties of these formulas are investigated. This paper also discuss the differences between the proposed distance measures. Finally, combined with satisfaction degree formula these distance measures are used in group recommendation. We further analyze the influence of the parameters of distance measures to the satisfaction degree and the influence of the parameters of satisfaction degree to the movie ranking.