Design And Implementation On Visualization Model For Analysis Of Literature Lending Based On Hadoop

For the past few years,especially since the 18 th National Congress of the Communist Party of China,under the guide of innovation-driven national strategy,in the background of the great age of mass entrepreneurship and mass innovation,internet + college informatization management has been a hot issue.The library,as a college department with great information data,with the gradual maturity of the emerging technology of mobile network,big data,cloud computing and IOT(Internet of Things),etc.and the increasing attention to library informatization in colleges,has realized the informatization management of more and more book resources,and made it take on a function-diversified and flowering situation.Great amount of data will be generated during the process of searching and consulting in libraries by the general teachers and students,and the data is the evidence through which we perform deep analysis of readers' reading and searching interest.It's worth noting that readers' diversified searching,interest analysis and intellectualized demand for books' application and recommendation bring the library huge challenges as well.On the one hand,the library intellectualized hardware facilities of single servers with the addition of disk array framework are unable to fit the need of the searching,analyzing and storing of great amounts of data.Moreover,the hardwares mentioned above have the defect of higher cost and the disadvantage that they are difficult to adapt to the big-data-era demand;On the other hand,at present,the method of analyzing college library borrowing circulation data still rests on the analysis of all kinds of single index objects,and the analysis is not comprehensive enough.At last,the current readers' interest analysis only arrives at a statistical result rather than the data which can further convert the statistical result into the intuitive guide to college library's other work such as purchasing.In order to solve the above problems,through performing data analysis and field research in the libraries ofan University in Changchun and part provincial colleges around it,integrating with big data theory and my tutor's research direction and relevant horizontal subject,based upon the deep research on Hadoop big data technology,setting up interest analysis and orientation model can play a certain reference role to the study of college library informatization construction.This dissertation performs the study mainly from the following four aspects: the first is to apply Hadoop big data technology and C# language data analysis technology to the work of reader's interest and orientation analysis,and,at the same time,in order to solve the high cost problem of big data storage and computation,this thesis chooses low cost Microsoft Azurez as server group used to construct Hadoop data platform;the second is to analyze libraries' reader borrowing search log(visiting origin of electronic book resources)through taking advantage of NoSQL distributed database and HBase database.Monitoring and optimizing of the using information of library electronic resources will be realized by means of log analysis;The third is to construct document recommendation model framework through the excavation of the historical data of readers' document borrowing;The fourth is to convert book recommendation list generated from document recommendation model framework,and finally to generate purchasing list for the collecting and editing department.This dissertation realizes the following three aspect innovation: the first one is to apply the current popular Hadoop big data analysis technology to the work of readers' interest analysis and orientation in traditional college libraries,to realize the construction of document borrowing analysis group platform in the low-cost computers,and to solve the distributed storage problems of great amounts of document borrowing information by fully using Microsoft cloud platform.The second is to introduce C# programming language into Hadoop framework in the aspect of data analysis,to provide efficient and intuitive data analysis scheme for the construction of interest analysis and orientation model,to perform big data analysis and comparison of readers' borrowing historical data,and finally to realize individualized recommendation for readers,and then to provide strong decision support for the construction of library collection and the structural adjustment of library book resources.The third is to realize individualized recommendation books,and to provide reference for purchasing books by libraries' book purchase department.