Design And Implementation Of Hierarchical Clustering And Visualization System Based On Web

With rapid development of informatization and increasing popularity of Internet applications in our society, there is a growing amount of data to be analysised and processed, and also there are more two-dimensional matrix data that reflect the relationships between things. Hierarchical clustering analysis and visualization of these data is an important means of mining the hidden information in these two-dimensional matrix data. C urrently there already exist systems and platforms for hierarchical clustering analysis and visualization of the data. But they have deficiencies in integration of the two parts, versatility, reprocessing of results, data storage and sharing. Therefore this paper proposes such a system based on web.This paper first analyzes the requirements of the system from the aspects of input file type, user group, functional and performance requirements. Then describe in detail the design and implementation of hierarchical clustering and visualization system based o n web. The system is designed into user module, data module, image module, management module. The system processes the uploaded data with hierarchical clustering and plots the results into heat map as requirements. Image generated is displayed through image browsing framework in web page. We can add tags, select area and enlarge to view detail or recluster to see the result. The system has realized the connection with Dropbox. It also supports image sharing to other web applications. The system relies on LAMP architecture. Data clustering is implemented by R. Image displaying is based on IIPImage. User interface is achieved using HTML5 and Ajax. At the end, the actual deployment and function and performance testing are introduced. After the testing, the system has achieved the expected goals.The system's main novelty lies in the fact that, we introduced social elements, realized the connection with the cloud platform, image browsing and data processing based on high-definition image display frame work, image sharing, etc. The system has been put into trial using and has provided support in the research of biological gene processing, and will play an important role in the research of related fields in the future.