Continuous Classes in Rule Induction: Empirical Comparison of Two Approaches
نویسندگان
چکیده
This paper introduces modiication of two existing rule-inducing ML algorithms; they are now capable of processing continuous classes. The rst one is the CN4 algorithm, a large extension of the well-known CN2. It invokes its discretization procedure and also deals with continuous classes within the inductive process itself, i.e., 'on-line'. The other ML system, KEX (Knowledge EXplorer), treats numerical attributes by its preprocessor, that is, they are discretized 'oo-line'; the continuous classes are processed within the inductive algorithm itself. Experimental results show a comparison of KEX and CN4 on some real-world as well as artiicial data.
منابع مشابه
Debt Level and Fiscal Rules in Iran According to Budget Structural Reform
Oil-exporting countries often escalate macroeconomic fluctuations by adopting cyclical fiscal policies. Empirical evidence shows that the main reason of instability in oil-exporting countries is the poor management of oil. In this study, after introducing a small open macroeconomic model with a dual managed floating exchange rate regime and estimating it using quarterly macroeconomic data of Ir...
متن کاملCombining rough sets and rule based classifiers for handling imbalanced data
The paper presents two rough sets based filtering approaches combined with rule based classifiers suited for handling imbalanced data sets, i.e., data sets where the minority class of primary importance is under-represented in comparison to the majority classes. We introduced two techniques to detect and process inconsistent majority cases in the boundary between the minority and majority class...
متن کاملNew Approaches to Analyze Gasoline Rationing
In this paper, the relation among factors in the road transportation sector from March, 2005 to March, 2011 is analyzed. Most of the previous studies have economical point of view on gasoline consumption. Here, a new approach is proposed in which different data mining techniques are used to extract meaningful relations between the aforementioned factors. The main and dependent factor is gasolin...
متن کاملExtending rule based classifiers for dealing with imbalanced data
Many real world applications involve learning from imbalanced data sets, i.e. data where the minority class of primary importance is under-represented in comparison to majority classes. The high imbalance is an important obstacle for many traditional machine learning algorithms as they are biased towards majority classes. It is desired to improve prediction of interesting, minority class exampl...
متن کاملComparison between induction effect with pulsatile and continuous oxytocin administration on outcomes of pregnancy in the pregnant women referring to Tabriz 29 Bahman hospital in 2006
Abstract Background: Insufficient uterine activity or hypotonic uterine contraction is a common reason and the cause of abnormal labor progress that can be corrected with oxytocin. Realizing this fact that undue prolongation of labor may contribute to prenatal morbidity has resulted in using oxytocin infusion in different kinds of ineffective uterine contractions. Materials and methods: In th...
متن کامل