Remote Sensing Cloud Users' Behavior Analysis Based On Clustering Algorithms

In recent years,with the rapid development of cloud computing technology and Earth observation technology,a remote sensing cloud service platform based on cloud computing technology and providing remote sensing application services is becoming more and more popular.However,with the change of the number and demand of users and the increasing volume of remote sensing data,the remote sensing cloud service platform is no longer satisfied with simply providing users with remote sensing application services.It is very important to provide users with diversified and customized high-quality services.In order to improve the service quality of remote sensing cloud platform,it is necessary to understand the behavior law of remote sensing cloud users interacting with each remote sensing service system.Therefore,the research of remote sensing cloud user behavior is carried out which was based on clustering algorithm in this paper.This paper mainly introduces the research work from two aspects of algorithm and application,and the specific research contents and contributions are summarized as follows:First of all,this paper investigates the user behavior analysis methods and related technologies in cloud computing environment and puts forward the research purpose and research content according to the research background of the subject.Based on the traditional data preprocessing technology,the user behavior data is cleaned and characterized,and finally provides accurate data analysis basis for user behavior analysis.Secondly,combining the semantic understanding of remote sensing data and products,the K-means clustering algorithm based on optimization achieves the behavioral feature cluster analysis of user behavior of remote sensing data services and remote sensing information products services.Solved the traditional method of feature analysis of the status quo,more accurately and accurately tap the statistics of different types of users of remote sensing data and product production requirements,in order to achieve accurate remote sensing data and product prediction to provide the basis.Then,based on the cluster analysis,the user data access and product production forecasting methods based on ALS matrix decomposition are designed to realize the estimation of remote sensing data and products respectively.Finally,the personalized and large-scale active production services of data services and product services in remote sensing cloud service systems were realized.Finally,in order to improve the quality of customer service and meet the timeliness of mass data loading and improve the production efficiency,the recommendation process for remote sensing cloud active service system based on user behavior analysis and prediction is designed.