Feature Transformation And Subset Selection
Huan Liu, Hiroshi Motoda · IEEE Intelligent Systems and their Applications · 1998
As computer and database technologies constantly advance, human beings rely more and more on computers to accumulate data, process data, and make use of data. Machine learning, knowledge discovery, and data mining are some of Arti cial Intelligence (AI) tools that help mankind accomplish those tasks. Researchers and practitioners realize that in order to use these tools e ectively, an important part is pre-processing in which 1 data is processed before it is presented to any learning, discovering, or visualizing al-gorithm. In many discovery applications (for example, marketing data analysis), a key operation is to nd subsets of the population that behave enough alike tobeworthy of focused analysis [1]. Although many learning methods attempt to either select/extract or construct features, both theoretical analyses and experimental studies indicate that many algorithms scale poorly to domains with large numbers of irrelevant and/or redundant features [6]. All the evidence suggests the need for additional methods for this purpose. Feature transformation and subset selection are some frequently used techniques in data pre-processing.