Data-mining by probability-based patterns
Mas'ood Karegar, Ayaz Isazadeh, Farzaneh Fartash, Taha Saderi, Ahmad Habibizad Navin · 2008
In this paper a new method is suggested for designing patterns in data-mining. These patterns are designed using probability rules in decision trees and are cared to be valid, novel, useful and understandable. By using the suggested patterns in data-mining, the system gets efficient information about the data stored in its data-bases and uses them in the best planning for special objectives.