Research On The Group-buying Store Clustering Model Based On Improved Customer Evaluation Of Emotional Inclination And Sales

Under the “Internet Plus” background,the number of online shopping consumer groups has already grown in size,gradually establishing a sound online shopping market operation mechanism.Group purchase industry is a product of the development of e-commerce industry to a certain extent.As a new type of e-commerce marketing model,it not only has a significant impact on the traditional retail industry,but also subtlely influences all aspects of people's lives.As the size of group purchases continues to grow and the transaction volume increases,commodity evaluation data will also be generated in large numbers.Therefore,data mining on the evaluation information of merchants will play an active guiding role in group purchase marketing.The cluster analysis of store evaluation and sales volume can improve the research of group purchase through data mining technology,and it is also an inevitable requirement for the rich industrial structure in the field of e-commerce group purchase.At present,Beijing has a relatively perfect development in the field of group purchases,and the catering sector also occupies a relatively large proportion of sales in group purchases.Therefore,it is more representative of Beijing Meituan catering area as a research object.This article analyzes and researches 32,000 evaluation data of 800 group-buying restaurants in Beijing,adopts the K-Means model to conduct cluster research on the sentiment trends and sales of group-buying merchants,and uses the support vector machine technology to do sentiment analysis on shop evaluation texts.And continuously predict the quantified sentimental value for use in subsequent store cluster analysis.In this paper,the text sentiment classifiers are constructed based on the support vector machine and the sentiment dictionary respectively,and the sentiment tendencies of the US group store evaluation texts are analyzed respectively.After the sentiment classification of evaluation texts based on the support vector machine,a continuous forecast of sentimental value was performed for the specific US group merchants,and the emotional value was adjusted to obtain the value of the store's sentimental value.When the sentiment classification was performed based on the sentiment dictionary,the group buy was constructed.Catering field emotional word dictionary.According to the comparison and analysis of the performance of the two classifiers,the SVM model was finally adopted as the sentiment classifier of this article to conduct sentiment analysis on the group-buying shops.The K-Means model is used to cluster the merchants in the field of group purchase to further study the relationship between various stores' emotional tendency values and group purchase sales.This paper evaluates the clustering effect by the contour coefficient and finds that when the initial k=4,the contour coefficient is the largest and the clustering effect is the best.There are four categories of shops in the group of catering services: high ratings-high sales(422),high ratings-low sales(167),low ratings-high sales(96),low ratings-low sales(106).This article further analyzes the emotional tendencies and sales data of the four types of group-buying stores,sums up the data characteristics of each group-buying shop,and proposes a targeted group-buying marketing strategy.