Tourism Data Mining And Research Based On Biclustering Method

Biclustering Algorithm has been well developed in the field of gene expression data analysis since it was put forward by Cheng and Church in 2000.However,its application in other fields remains to be explored,especially in the field of tourism big data.Tourism big data in recent years as a hot topic,many are used to predict the peak travel periods,the number of customers in each tourist attractions,among other.With the improvement of people's life quality,the demand for the tourism quality will be higher and higher.Therefore,independent travel and individualized tourism are chosen by more and more people.Therefore,it is a very meaningful study to classify tourists and develop personalized travel routes and service for different types of tourists.This paper is applying the biclustering algorithm into the research of tourism big data.The results of user cluster can be used in developing potential tourists,making individualized travel routes as well as predicting the heat of a tourist attractions.This paper uses Python mining on the data of network users of a tourism website.Then clean the content stored by Excel.Using the ICTCLAS to separate the texts.Using TF-IDF algorithm to extract keywords,and simple word frequency statistics are used for these keywords.And taking the keywords appear frequently as keywords matrix variables,using the user as the object,establish the user-key word matrix.Then do the biclustering algorithm to the matrix.Using biclustering algorithm to cluster the matrix.By comparing the selected functions of different clustering algorithm,choose an algorithm fitting the high dimensional data by the requirement of clustering results,find out a suitable algorithm for high-dimensional tourism data.Results of data analysis show that the biclustering algorithm has a good clustering effect for the classification of tourist data.