Application of K-means clustering of inbound visitors to China

Abstract

The number of inbound visitors received by China is among the top in the world, which makes China's inbound visitor market very huge. Obtaining sub- visitor markets by segmenting the visitor market is beneficial to better meet the needs of visitors, helps travel agencies make market decisions and develop markets. This paper classifies inbound visitor of China and get the sub-visitor markets by K-means clustering method. Using K-means clustering method to classify visitor samples requires collecting data on different attributes of the sample. This paper divided total visitor attributes into information attribute and behavior attributes. Based on those two types of attributes this paper use K-means clustering to cluster three inbound visitor sample sets, which are total sample set (contain all the information attributes and all the behavior attributes), information sample set (contain each one of the information attributes and all the behavior attributes), behavior sample set (contain all the information attributes and each one of behavior attributes) to get clustering result. The clustering results are sub-visitor markets and provide appropriate policy suggestions to the tourism agencies basis the characteristics of each sub-visitor market. After clustering analysis, it is found that China's inbound visitors can be divided into 5 types, each of which has distinct characteristics. For example, the international student cluster is composed of a group of low-income young people whose average daily consumption is very low and mainly used for purchasing essential daily necessities. One of their most important characteristics is that the number of entries is relatively frequent and the stay time for each entry is very long.