| With the widespread use of location-based social platforms,more and more consumers have become accustomed to browsing online reviews to make their own consumption decisions.However,most of the existing relevant studies believe that consumers are completely rational people who have all the information.In real life,consumers’ actual decision-making process is usually affected by their own risk preference and external environment,which is manifested as bounded rationality.Based on this,it is of certain theoretical and practical significance to study the selection model of commercial interest points on the basis of fully considering consumers’ risk preferences,so as to provide correct guidance for consumers to make consumption decisions in the process of online consumption.In this thesis,an algorithm model is built based on the online review data of commercial interest points.Combined with the unique temporal and spatial characteristics of commercial interest points,an analysis of consumers’ risk preference behavior is introduced.Based on the prospect theory framework,a personalized selection model of commercial interest points and a group selection model of commercial interest points are constructed.Firstly,This thesis begins by summarizing the current relevant theoretical basis and technical research,including the selection of business interest points,the characteristics of consumer risk preference and online review information,and sorts out the theories and technologies that can inspire and guide the construction of the selection model below.Secondly,a personalized selection model of business interest points was established,and the power law distribution model and access popularity were combined to analyze the characteristics of consumer preference based on spatio-temporal factors.Web crawler technology was used to collect online comment information on commercial interest points and pre-process it.The emotional strength analysis method was used to quantify the scale value.Taking the expectation level given by consumers as the reference point,the personalized selection model of business interest points is built based on the prospect theory,and the real data on the Yelp platform is used to verify the model.Finally,a group selection model of business interest points was established to describe the differences of members from the perspective of distance influence degree coefficient of each member and individual prediction score respectively.The check-in experience values of members in the group were obtained according to the number of historical comments.On this basis,combined with the characteristic analysis of preference fusion strategies,the selection of different preference fusion strategies under different situations was realized.The final group selection results of business interest points were determined for group members,and combined with the real data set on the Yelp platform,the validity of the above model was verified based on Python software.The results show that both the personalized choice model and the group choice model take into account the characteristics of consumers’ risk preferences,that is,the business interest point choice model under the framework of prospect theory can better simulate consumers’ choice and decision-making behavior in real life.The model proposed in this paper makes up for the gap in the current relevant research,provides a theoretical basis and decision support for consumers’ choice decisions,and provides a more scientific method for determining the optimal scheme. |
