Browsing by Author "Liao, SY"

Now showing 1 - 2 of 2

Data types generalization and transformation for data mining algorithms
(2000-06-01)
Jiang, MF
;
Tseng, SS
;
Liao, SY
;
資訊工程學系
;
Department of Computer Science
In most applications, knowing the characteristics of the data mining methods is usually the key factor to determine the kind of data mining methods should be applied. In this work, we first survey and analyze six kinds of data mining algorithms and four kinds of data forms, and then find out the relations between the mining algorithms and the data types. According to the relations, the suitable generalized data types will be created and is then used to transform the generalized data types of data sources to the suitable one for the selected mining algorithm. With the proposed preprocessing work for data mining algorithms, users can select appropriate mining algorithm just for the goal of application without considering the data types. Finally. two applications for e-mail management and e-mail log analysis are proposed to show the practicality of our method.
Two-phase data types transformation framework in data mining
(2001)
Jiang, MF
;
Tseng, SS
;
Liao, SY
;
Chen, WC
;
資訊工程學系
As we know, the data processed in data mining may be obtained from many sources in which different data types may be used. However, no algorithm can be applied to all applications due to the difficulty for fitting data types of the algorithm, so the selection of an appropriate mining algorithm is based on not only the goal of application, but also the data fittability. Therefore, transforming the non-fitting data type into target one is also an important work in data mining, but the work is often tedious or complex since a lot of data types exist in real world. Merging the similar data types of a given selected mining algorithm into a generalized data type seems to be a good approach to reduce the transformation complexity. In this work, a two-phase data types transformation framework including merging and transforming phases is proposed. With the data type transformation framework, the user can select appropriate mining algorithm iterative and interactive for the goal of application without considering the data types.