|
Vol: 47(61) No: 2 / June 2002 Mapping Quantitative Attributes for the Association Rule Problem Luminita Dumitriu Dept. of Computer Science and Engineering, "Dunarea de Jos" University, str. Domneasca nr. 111, 6200, Galati, Romania, phone: 40-723-161314, e-mail: Luminita.Dumitriu@ugal.ro Keywords: data mining, association rules, data models. Abstract Mapping quantitative attributes has been solved in several ways, mainly grouping their values in ranges. In the absence of a priori knowledge, there is no criterion to evaluate the appropriateness of the mapping. We have devised an approach that allows the user to build partial data models, based on some of the attributes in the database. For the attributes with indecidable mapping we are offering the user a data analysis tool that measures association degrees between the values of a new attribute to an existing, partial data model. We have defined some measures of association that can quantify the appropriateness of some mapping to an existing data model. Using the resulting values, the user can easily select the set of attributes that lead to detecting the mapping. References [1] Agrawal, R., Imielinski, T., Swami, A.: Mining association rules between sets of items in large databases. In Proc. of the ACM SIGMOD Conference on Management of Data, 207-216, Washington D.C., May 1993. [2] Aumann, Y., Lindell, Y.: A statistical theory for quantitative association rules. In Proc. of KDD \'99, 261-270, San Diego, C A, 1999. [3] Fukuda T., Morimoto, Z., Morishita, S., Tokuyama, T.: Mining Optimized Association Rules for Numeric Attributes. In Proc. of the 15th ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database System, 182-191, 1996. [4] Kuok, C. M., Fu, A., Wong, M. H.: Mining fuzzy association rules in databases. The ACM SIGMOD Record, Vol 27, No. 1, 41-46, 1998. [5] Miller R.J., Yang, Y.: Association Rules over Interval Data. In Proc. of ACM SIGMOD, vol 26, 452-462, 1997. [6] Rastogi, R., Shim, K.. Mining optimized association rules for categorical and numerical attributes. Technical report 0112370-970217-03, Bell Laboratories, Murray Hill, 1997. [7] Srikant, R., Agrawal, R.: Mining quantitative association rules in large relational table. In Proc. of the ACM SIGMOD Conference on Management of Data, June 1996. |