Research On Tourism Information Mining Based On Plotting Data

With the development of location based services,geographic information system and mobile terminal technology,people increasingly interested in the information released with position attribution using a mobile phone,and produced a large amount of plotting data.Plotting data naturally has geographical location attribute and time attribute,and has huge data quantity,various forms of expression,complex data structure and extensive data source.Plotting data is the data generated by people using the Internet to express their attitudes and opinions about certain things.It is closely related to people's lives.Therefore,this kind of data has the great value of mining.This paper takes tourism as an example to excavate tourist information in plotting data.The purpose of this study is to obtain tourist information from the plotted data,and use these tourist information to help visitors plan travel and assist the decision-making of tourism departments.Because the plot data contains many aspects of content,it is necessary to obtain tourism data.Firstly,the text information in the plotting data is used to chinese word segmentation,and then the data are divided into tourism subject data and non-tourism subject data by using the naive Bayesian algorithm.Finally,the data of the tourism subject data are excavated and obtained the tourist information.This article obtains the traveling information from two aspects: the hot spot mining and the tourism abnormal event excavation.Hot spot mining uses the spatial clustering algorithm to cluster the tourism plotting data so as to achieve the purpose of catching hot spots.This paper studies the Kmeans algorithm as the representative of the partition based spatial clustering method and DBScan algorithm as the representative of the spatial clustering method based on density.Because the Kmeans algorithm randomly set initial clustering center,the clustering results easy to fall into the local optimal solution,and the algorithm requires the user to input parameter number of categories K,it is difficult for users to give appropriate values when they are not familiar with the data distribution,and ultimately lead to users can not get better accuracy of clustering results.In this paper,based on the idea of density clustering,Kernel-Kmeans algorithm is proposed to solve these defects,and the results of higher accuracy are guaranteed.By using this method,we can get information about hot spots and scenic spots.Tourism abnormal event refers to the abnormal amount of tourism plotting data.By comparing the amount of daily tourist plotting data in a scenic spot for a period of time,the amount of tourist plotting data is detected far more than the usual date.Refer to relevant information to see if the site held events on that date.This paper detects the world Internet Conference and Wuzhen Drama Festival in Wuzhen scenic spot by abnormal detection of box diagram.This paper develops spatial data mining and knowledge service system in order to display the excavated tourism information.The system mainly displays the results of hot spot mining through HeatMap display technology,showing the hot spot information of tourist attractions,hot spots of provinces and the hot spots of national tourism.The system also displays the scenic spots information and the scenic abnormal events through the HighCharts technology.