|
With the success of research and development of the theory of data mining research, in-depth and all kinds of data mining software (tools), data mining technology to be more widely used in various fields. The successful application of these areas has brought great inspiration to the library. Produce various types of data in the library of modern management, automation system also contains a large amount of data. According to the characteristics of these data, targeted theoretical research, choose suitable for the the Library application of data mining software, and how to use these software mining the library meaningful results for readers to access the library's purpose and trends to understand the reader's interest and demand, improve the quality of service to meet the needs of readers, enhance reader satisfaction, give readers a better service for library managers to provide decision-making reference views is a problem worthy of study. First, according to the needs of library data mining, a lot of data collected, including: paper questionnaire design and data management, the university library data network to collect and collate the data in the \extraction and finishing. The paper questionnaire covers the construction of library resources, use and service conditions, the data include: the ease of access to information, and much-needed increase holdings extent and degree of familiarity with the various departments of the library; network data collection of the University Library including the: Library hardware facilities, human resources, library collection, and opening hours; the universal integrated library system in the data extracted include: readers information, book lending records information and Readers record information; data content coverage than wide. Then, a detailed, in-depth exploration of data mining technology in library management application and implementation process. The questionnaire collected data and network data collected were clustering algorithm based on density and density up to provide decision support to improve the overall level of service to the library by discovering meaningful clusters. Clementine association analysis of the extracted data in general-purpose integrated library system, part of the data on paper questionnaires a quantitative association algorithm based on distance analysis, according to the the experimental association rules identify between books and books, readers and books association between interdisciplinary, do book recommendation services. Finally, summed up the library data mining, decision trees, respectively, for the data layer, technology layer and the decision-making level. Clustering results given Readers demand, the demand for research projects, to improve the quality of library services and the to good library human resources planning and other aspects of the decision-making proposals, according to the the associated results given book purchasing, library shelving, discipline construction aspects of the decision-making recommendations.
|