Studies On Decision Behavior And Consensus Reaching In Large-Scale Group Decision Making With Heterogeneous Preference Structures

Group decision making,which is consensus reaching process implemented by a group of decisnion makers or experts in terms of some certain criteria or objectives for a decision making question,has become an important branch of management science researches for its profound theoretical background and broad application.It is an interdisciplinary research field assoiated with organizational behavior,decision science,social psychology,information techinology,behavior science and so on.With the widespread internent applications and progress of big data techinology,the participants of group decision making ranging from serveral experts to large–scale decision makers,and the large–scale group decision making become a main branch of group decision making.In large-scale group decision making,different decision makers who have different knowledge backgrounds,social status,habits and demands always provide the heterogeneous preference relations.Moreover,non-cooperative behaviour also exists in subgroup or individual preference.Therefore,the coordination of large groups of decision makers with different views and conflicting views and consensus reaching of collective opinion are more complex and chanlleage issue compare to traditional group decision making with small group of participants.For example,decision-making in social networks,commodity evaluation in e-commerce,democratic decision-making in large groups,supplier selection in the whole industry supply chain,and so on.At present,the problems of large-scale group decision-making are becoming urgent topic with widely practical applications.However,considering different preference forms and non-cooperative behavior in large-scale group dicision making,new tools need to develop to deal with large-scale group participants such as the detection and management of non-cooperative behavior,intergrating different preference structures and classification of decision makers.Motivated to these qurstions,this paper will study the consensus reaching mechanism with heterogeneouspreference relations and non-cooperative behavior,which involve intergrating different preference relations,data mining algorithms for largr–scale group decision making and consensus reaching models,and so on.The main research contents include three parts shown as follows:(1)Researching on intrgrating heterogeneous preference relations.Firstly,this paper briefly overviews the related concepts,the natures and the theorems of utility value,preference ordering,multiplicative preferrece relation and fuzzy preference relation,which are the main objectives of this paper.Then,constructs a derivation method for priority vector of fuzzy preference relation,and applies to ranking of 1544 Go plays on the basis of cosine similarity measure.The comparision of existing method is also be shown in order to prove effectiveness of proposed methods.Lastly,this paper proposes a similarity-based optimization model to intergrate four different preference forms and incomplete preference relations.The results compared to other existing methods are also provided in this section.(2)Researching on data mining technologies used to large-scale group decision making.Firstly,this paper briefly reviews related concepts of large–scale group decision making including research questions,main features and research methods.Then,proposes the measures of distance among different preference forms and introduces clustering algorithm for large-scale group decision making.Secondly,this paper develops a multi-criteria quadratic programming model for imbalanced data.This method can be used to forecast minor decision maker behavior in large-scale group decision maker.(3)Researching on the methods for consensus reaching.Firstly,this paper builds a consensus reaching model based on multiplicative preference relation.Then,this paper constructs a method for detecting non-cooperative behavior based on clustering algorithm.Secondly,a consensus reaching mechanism is proposed include aggregation,clustering,modification,detection and management processes.Lastly,the model is applied to selection of aiding objection in ending poverty project in China.In short,studies on decision behavior and consensus reaching in large-scale group decision making with heterogeneous preference structures are important research content in the field of group decision making.Moreover,these problems have also become a reality of real-world application.However,there also is a gap build a complete consensus reaching model to handle non-cooerative behaviou and heterogeneous preference forms.Therefore,this paper introduces consensus reaching in large-scale group decision making with data mining technologies,and establishes a complete framework for consensus reaching.The further research in this paper can not only promote the development of large-scale group decision support system,enriched methods for the group decision-making and techniques,but also has a good guiding significance on real-life management application.In addition,this paper also puts forward the future of large-scale decision-making prospects and proposes several further research issues in this area.