The Application Of Data Mining Technology In Archival Information Management System

Data mining is searching for hidden knowledge by the algorithm based on a lot of dates,it has been widely used in many data-intensive industries.Archives refers to preserve the value of the original records of people in various social activities in various forms of direct form,in our archives computer management sector over the years formed a large number of electronic data,but the use of these data is still relatively junior,only statistics and analysis.In view of this situation,this paper attempts between the data warehouse and data files and user,between archives and the various units and file archiving mining technology archives information management work,through the analysis of the use of databases and archives archives archive database revealed relationship,resulting in a deeper level of play value archive data in the new era.The main aspects of the article from the archives of data warehouse design,data ETL,establish multidimensional data model,data mining algorithm selection,data mining model application discusses the practice of data mining technology and the combination of archives,archives of data mining to achieve the basic process.First,build a data warehouse for detailed needs analysis to determine the theme and users to determine the data warehouse dimension tables and fact tables based on an existing database file operations,and eventually create a file using a star schema data warehouse;then will archive the data ETL After the original data source extraction,transformation,and loading into the data warehouse,this process is mainly for archival data preprocessing and clean;then re-use OLAP cube tool to archive query multiple perspectives and multiple views to better understanding archival data;Finally,based on archival data warehouse,data mining based on needs analysis,use SQL Server Analysis Services tools and select the appropriate data mining algorithms to archive data analysis and processing,and the results were analyzed.The core of full-text data mining technology practices in the file information management.