J. Deogun, V. Raghavan, A. Sarkar, H. Sever (1997), Data mining: Trends in research and development, In: Lin and Cercone [14], pp. 9--45.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....and others. The methods of data mining in very large data bases using rough sets recently have been proposed and investigated. There have been done a substantial progress in developing rough set methods for Data Mining and Knowledge Discovery (see methods and cases reported in e.g. in [3] 4] [5], 9] 10] 14] 17] 18] 21] 29] 30] 31] 41] New methods for extracting patterns from data (see e.g. 12] 19] 17] 11] 27] decomposition of decision tables (see e.g. 19] as well as a new methodology for data mining in distributed and multiagent systems (see e.g. 28] ....

J. Deogun, V. Raghavan, A. Sarkar, H. Sever (1997), Data mining: Trends in research and development, In: Lin and Cercone [14], pp. 9--45.

.... Not in family, Other relative, Own child, Unmarried and cap gain 5119 and cap loss 820.5 then class: 50K [163, 51] 8. if educatn Doctorate, Prof school and relation Not in family, Other relative, Own child, Unmarried and cap gain 5119 and cap loss 1820.5 then class: 50K [2, 11]. The first rule, confirmed by the most of cases from the training data, indicates that adults with lower level of education and rather low values of capital gain and capital loss are most likely to have salaries less than 50,000. The third rule shows that adults that are married (relationship ....

....of companies was drawn from the DISCLOSURE Worldscope Global data file and Polish Stock Market. The sample consists of 279 firms from France, Germany, U.K. and Poland. The ProbRough system generated the rough classifier that consists of four decision rules: 1. if ROA 0. 393 then D = Poland [2, 5, 3, 60] 2. if 0.393 ROA 5.825 and CR 1.518 then D = Germany [8, 46, 4, 0] 3. if ROA 5.825 then D = U. K. 35, 12, 53, 0] 4. if 0.393 ROA 5.825 and CR 1.518 then D = France [28, 10, 13, 0] The generated rules can easily be illustrated in the plane determined by the CR and ROA attributes ....

Deogun, J. S., Raghavan, V. V., Sarkar, A. and Sever, H. (1997). Data mining: trends in research and developments. In: Lin, T. Y. and Cercone, N. (eds), Rough Sets and Data Mining: Analysis of Imprecise Data, Kluwer Academic Publishers, Boston/ London/ Dordrecht, pp. 9-45.

....to only a subset of data, altering existing data or introducing additional (spurious) data. They contend that the application of such policies is most effective in the context of specific learning tasks. These tasks include classification, estimation, clustering, characterization and association [2, 3, 4]. Of special interest to the current work are the classification mining algorithms, which have the potential to disclose sensitive information whenever a database contains both sensitive and non sensitive data [1] Specifically, our current work has focused upon the need to develop security ....

....of the proposed security policies. Section seven presents an outline of future research projects. 1. CLASSIFICATION MINING AND SECURITY The goal of classification mining is to discover patterns that classify objects, or tuples in the context of the relational data model, into predefined classes [2, 3, 4]. This goal is achieved, in part, through the successful completion of three specific tasks. The first task is the selection of an attribute from the given relation. The selected attribute is typically referred to as the decision variable since its purpose is to partition tuples into disjoint sets ....

J. Deogun, V. Raghavan, A. Sarkar, and H. Sever, "Data Mining: Trends in Research and Development," in Rough Sets and Data Mining: Analysis for Imprecise Data, Boston, MA, pp. 9-45, 1996.

....to only a subset of data, altering existing data or introducing additional (spurious) data. They contend that the application of such policies is most effective in the context of specific learning tasks. These tasks include classification, estimation, clustering, characterization and association [1,3,4]. Of special interest to the current work are the classification mining algorithms, which have the potential to disclose sensitive information whenever a database contains both sensitive and non sensitive data [2] Specifically, our current work has focused upon the need to develop security ....

....the work presented in this paper and presents an outline of future research projects. 2 Classification Mining and Database Security The goal of classification mining is to discover patterns that classify objects, or tuples in the context of the relational data model, into predefined classes [1,3]. This goal is achieved, in part, through the successful completion of three specific tasks. One of the required tasks is the selection of an attribute from the given relation. The selected attribute is typically referred to as the decision variable since its purpose is to partition tuples into ....

J. Deogun, V. Raghavan, A. Sarkar, and H. Sever, "Data Mining: Trends in Research and Development, " in Rough Sets and Data Mining: Analysis for Imprecise Data, Boston, MA, pp. 9-45, 1996.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC