Abstract— The term data mining has been the oldest yet one of the interesting buzzwords. Many organizations often underutilize their already existing databases. There is a need to mine information and interesting patterns from these databases. The focus of the current research is to apply data mining on a library management system. Data mining is usually done on a data warehouse or a data mart. It incurs various cost factors like software, hardware, maintenance and experts. The objective here is to study how the real-time data stored in database can be turned informative without setting up a separate data warehouse. The main emphasize is on understanding the problem perspective, competing objectives and constraints and generating a model for information extraction from the real-time library database using ARM (Association Rule Mining) mining technique. As SQL (Structured Query Language) can also be used for mining data instead of using specialized data mining algorithm [1], the study also compares SQL based mining with ARM. The results shows that association rule mining performs better than SQL based mining as type of pattern to be extracted can be controlled much effectively in ARM as compared to SQL because of the parameters (support and count) used in the data mining algorithm. Algorithms are implemented using SQL and MATLAB (Matrix Laboratory) Tool - ARMADA.
Index Terms— ARM, ARMADA, data warehouse, data mart.
M. Bansal is working with Department of Information Technology, Shaheed Udham Singh College of Engineering & Technology, Tangori, Punjab , India (e-mail: mrmanubansal@gmail.com).
M. Kaur is with University Institute of Engineering and Technology, Panjab University, Chandigarh, Punjab, India (e-mail: mandeep@pu.ac.in).
[PDF]
Cite: Manu Bansal and Mandeep Kaur, " Analysis and Comparison of Data Mining Tools Using Case Study of Library Management System," International Journal of Information and Electronics Engineering vol. 3, no. 5, pp. 466-469, 2013.
