基于模糊一致性调整算法的推荐系统评估研究

Research on Evaluation of Recommendation System Based on Fuzzy Consistency Adjustment Algorithm

考虑到一致性调整和排序权重向量确定方法是犹豫模糊语言群决策中的2个重要课题,构造基于一致性调整算法的犹豫模糊语言群决策模型。具体来说,该模型首先运用对数最小二乘法计算犹豫模糊语言偏好关系(hesitant fuzzy linguistic preference relation,HFLPR)的排序权重向量;其次,将一致性指数最小的语言偏好关系对应的排序权重向量作为HFLPR的序权重向量;随后,基于一致性调整算法建立了一种收敛的犹豫模糊语言群决策模型。该模型既没有改变原始犹豫模糊语言术语集,又可以尽可能地避免信息的损失;最后,将建立的模型应用于推荐系统的优选过程。实验表明,构建的群决策算法更合理有效。

Considering that consistency adjustment and weight vector determination are two important topics in hesitant fuzzy linguistic group decision making,a hesitant fuzzy linguistic group decision making model based on consistency adjustment algorithm is constructed.Specifically,the logarithmic least square method is used to calculate the ranking weight vector of hesitant fuzzy linguistic preference relation(HFLPR).Secondly,the ranking weight vector corresponding to the linguistic preference relation with the smallest consistency index is used to the ranking weight vector of HFLPR.Then,based on consistency adjustment algorithm,a convergent hesitant fuzzy linguistic group decision-making model is developed,which is neither change original hesitant fuzzy linguistic term set,and can be used to avoid the loss of information as much as possible.Finally,the developed model is applied to the optimization of recommendation system process,the experiment shows that the developed group decision model is more reasonable and effective.