A Clustering Research Of The Catering Business Based On The Dianping

Using data mining technology to classify merchants is an important prerequisite for the understanding of their own situation and the formulation of the restaurant improvement plan.With the continuous improvement of the living standards of the residents,the proportion of service industry in the national economy has gradually increased,and the catering industry has been playing an indispensable role in the service industry.In the big data era of the Internet,the e-commerce platform has developed rapidly,and the catering industry can not only be satisfied with offline sales.Dianping,as an o2o form of consumer comment website,has emerged.At present,the number of merchants in Dianping has exceeded 1.5 million,and the number of restaurant merchants has been increasing day by day.It will be of practical significance to mine and analyze the data of online restaurant merchants.In the past,the research and application of merchant classification based on their scoring information were relatively few.In this paper,the crawler technology is used to obtain the evaluation information of 485914 restaurants on Dianping.Some statistical software such as SQL and Python,and common statistical methods are used to analyze and explore the basic information of Merchants.And then using Apriori algorithm,K-means algorithm and its improved algorithm to classify merchants.According to the classification results,different improvement schemes and suggestions are put forward for different groups of merchants.At the same time,some basis of dining selection are put forward from the perspective of consumers.To some extent,this article has achieved the breakthrough and innovation of the previous research.The main conclusions of this paper are as follows:1)In the exploratory statistical section.From the overall distribution of merchants.Beij ing,Shanghai,Guangzhou and Shenzhen have a relatively large number of businesses,as well as more restaurants dealing with bread and desserts,hot pot,Sichuan cuisine,barbecue,western food and snacks;In the correlation analysis,the amount of consumption,the quantity of comments and other indicators are almost independent of each other,and the correlation between each score is very high;In the distribution of commercial area,about 50%businesses belong to the mixed area,followed by the commercial entertainment area and residential area;The characteristics of eating habits,Local specialty cuisines are sold in large numbers in each city,and bread desserts,snacks,hot pots and barbecues are popular in almost every city.2)In the association analysis section.By using Apriori correlation algorithm,it is concluded that there is a correlation between the star mark and the environment,service and taste mark of the merchants.For restaurants with lower star mark,There is a greater probability that its environmental score is slightly higher than that of service and food.For some restaurants with higher star ratings,there is a high probability that the restaurant has a high taste score and a higher service score,and a lower probability of a higher environmental score.3)In the cluster analysis section.The K-means algorithm and its improved algorithm are used to cluster the merchants,and the merchants are divided into four categories:high quality class,middle and upper class,middle class,middle and lower class.and each category accounts for 7%,14%,3 8%and 41%of the total number of merchants respectively.Among them,the star level of the high quality restaurant is 45 to 50,the star of the middle and upper class restaurant is about 40,and the star level of the middle and lower class restaurant is about 35 and 30 respectively.It can be seen that the star level of the high quality restaurant and the middle and upper class restaurant is very high,but the proportion of the number is especially low,the total is about 20%,and the star level of the middle and low class restaurant is very low,but it accounts for more than 40%of the total number.To a certain extent,the quality of most restaurants in the current catering market is maintained at a normal level.There is also a need to make a number of improvements.