Business Data Analysis Based On Hierarchical Clustering

In 2018,with the development of IT industry and the breakthrough and innovation of mobile internet,the amount of global information data and data is growing rapidly and steadily.The arrival of big data provides more efficient and convenient leadership for e-commerce and various industries,and opens up a broader development and leadership field for it.With the increasing popularity of the era of big data in our life,the era of big data has greatly expanded the development and application of the Internet.In this era,we are in a "big data" era of explosive growth of data.Big data plays an indispensable role in social economy,military,culture,people's life,etc.The advent of big data and e-commerce The combination of development is the inevitable result of social development,which will lead us into a new situation of innovation and development.With the passage of time,the amount of data and information on the network has become more and more huge,but the ability of consumers to process information is very limited.It is difficult for consumers to screen and analyze a large number of data and information correctly.By collecting and analyzing network data,e-commerce can subdivide users into different consumer groups according to their consumption characteristics,and provide personalized services for users' consumption and preferences.This paper gives a brief introduction to the commonly used data format of web pages and makes a comparative study of the methods of parsing this data format.Combining with the popular data collectors in China,the actual commercial data which are used to analyze the data in this paper are collected.Secondly,the classical partition-based clustering algorithms such as K-Means algorithm and K-Medians algorithm,hierarchical clustering algorithm such as CRUE algorithm and BITCH algorithm,density-based clustering algorithm such as DBSCAN algorithm and OPTIC algorithm are discussed,and a simple example is given to illustrate the data partition of hierarchical clustering algorithm to be studied in this paper.Analysis.Finally,the collected data are analyzed by Kissmetric data analysis tools and hierarchical clustering analysis algorithm which are popular in foreign countries.By calculating the similarity(distance)of data points in hierarchical clustering analysis algorithm,the results are compared with the actual situation.The similarity between the two methods is high,which provides a certain reference value for practical commercial applications.At present,the research method has been put into use,and the results are very ideal,which greatly increases the sales of goods and brings considerable profits to enterprises.