Research On Large Group Multi Attribute Decision Making Method Of Tourism Experience Considering Information Quality

In the era of experience economy,people’s demands for product service experience are increasing rapidly.Tourism is a field where experiences occur frequently.Compared with physical products,users focus on basic functional satisfaction.In service products such as tourism,tourists pay more attention to whether the consumption process can bring about an improvement in personal pleasure.In a word,the quality of tourism experience in a destination has a direct impact on tourist satisfaction,and then affects destination decision making.There contains a lot of experience information in travel notes,which reflect the real destination quality of experience.Thus,this paper proposes a large group multi attribute method based on the quality of tourism experience extracted from travel notes considering the UGC information quality.First,considering that there are two kinds of information quality problems when UGC is used in decision making,namely,inconsistent attribute evaluation and incomplete evaluation scope,a new method for constructing decision subgroups is proposed in this paper.Firstly,this paper determines a unified evaluation attribute dimension for all users through attribute extraction and feature word mapping methods.Secondly,this paper constructs a sentiment computing rule based on domain sentiment lexicon to convert the evaluation terms into continuous values.Then the initial subgroup is obtained by clustering based on the similarity of attribute evaluation.Further,this paper develops a bi-objective optimization model based on the group attribute entropy indicator and group consensus degree indicator to modify decisionmakers in various initial subgroups,making the evaluation information in the subgroup as perfect as possible and the opinions are consistent.Second,considering the timeliness of travel information,a dynamic decision-making model is constructed.Firstly,time window is determined based on the similarity of attributes extracted from the travel notes in adjacent periods,and the time window weight is jointly defined based on the rule of forgetting curve and the similarity of attributes.For each time window,the corresponding subgroups are divided and the weight of the subgroup is jointly determined by the number of members and the degree of consensus of internal opinions.By integrating the subgroup evaluation,the static group decision-making matrix of different windows is obtained,and the dynamic decision-making model is constructed by combining the time weight.Finally,the destination ranking results are solved based on VIKOR method and weight of time and attribute.Finally,this paper verifies the destination ranking problem based on the experience information in travel notes.The proposed method tackles the limitation of incomplete information in group decision making on subgroup construction,and breaks through the traditional missing value processing methods that are difficult to apply to highly sparse evaluation matrices.At the same time,it effectively distinguishes the UGC decision reference value in different time periods based on the similarity of content.The method proposed in this paper can provide decision-making basis for the improvement of destination quality of experience and destination ranking.