Request PDF on ResearchGate | ChiMerge: Discretization of Numeric Attributes. | Many classification algorithms require that the training data contain only. THE CHIMERGE AND CHI2 ALGORITHMS. . We discuss methods for discretization of numerical attributes. We limit ourself to investigating methods. Discretization can turn numeric attributes into dis- discretize numeric attributes repeatedly until some in- This work stems from Kerber’s ChiMerge 4] which.

Author: Nenos Tojanris
Country: Nicaragua
Language: English (Spanish)
Genre: Music
Published (Last): 16 May 2013
Pages: 256
PDF File Size: 11.82 Mb
ePub File Size: 10.95 Mb
ISBN: 236-1-16591-295-7
Downloads: 57566
Price: Free* [*Free Regsitration Required]
Uploader: Kejora

In fact, when the number of some class increases, this class has stronger independence with intervals, and it has leader’s class status. In this problem we select one of the following as an attribute: Three Classes are available which are Iris-setosa, Iris-versicolorIris-virginica. But this criterion merely considered dependence between the most classes in the interval and the attribute, which will cause the excessive discretization and the result is not to be precise.

Moreover, there are two quite influential discretization methods which are the algorithms of the correlation based on information entropy and the algorithms of attribuhes correlation of Chi2 algorithm based on statistical method for supervised discretization. Therefore, compared with not increased, this time should have the same opportunity of competition and even should merge first these two intervals. This time, merged standard of extended Chi2 algorithm is possibly more accurate in computation.

Therefore, parameter as condition parameter can play a fair role: Check numdric ChiMerge powerpoint slides that visualizes the above algorithm. Enter your email address to subscribe to this blog and receive notifications of new posts by email. In the algorithm we adopt two operations.

An Algorithm for Discretization of Real Value Attributes Based on Interval Similarity

In algorithms of the series of Chi2 algorithm, expansion to is as follows: In other words, when is quite bigger thanvalue will increase degree of freedom not to change and probability of interval merging will be reduced. In view of an algorithm for discretization of real value attributes based on rough set, people have conducted extensive research and proposed a lot of new discretization method [ 5 ], one kind of thought of which is that the decision table compatibility is not changed during discretion.


It uses a user-specified number of intervals when initializing the discretization intervals. A good similarity measure should have the following characteristic: Net Algorithms Technology Errors! The chi 2 values are calculated for the revised frequency table, chi merge proceeds iteratively in this way merging two intervals at each stage until the chi 2 for the remaining pairs of intervals are greater than the threshold value and the number of intervals is less than the maximum number of intervals, hence no further merging of intervals is possible and the discretisation is complete.

Discretization algorithm of real value attributes actually is in the process of removing cut point and merging adjacent intervals based on definite rules.

From Table 4we can see that under 1-V-1 classification method the predictive accuracy with SIM algorithm is higher than that of extended Chi2 algorithm and Boolean discretization algorithm except for Breast and Pima datasets.

ChiMerge discretization algorithm November 2, Dicsretization in fact, adjacent two intervals with the bigger difference of class distribution and the greater number of classes should not be first merged. Chkmerge of distribution with different degrees of freedom.

So you could probably that the code below will compile only using Visual Studio and. Discretization algorithm for real value attributes is of very important uses in many areas such as intelligence and machine learning.

It attributws unreasonable to merge first adjacent two discretjzation which have the maximal difference value. Discretization of real value attributes is an important method of compression data and simplification analysis and also is an indeterminable in pattern recognition, machine learning, and rough set analysis domain. Therefore, even if, we still have such situation: The related theory analysis and the experiment results show that the presented algorithm is effective.


Huang has solved the above problem, but at the expense of very high-computational cost [ 9 ]. The smaller the value is, the more the similar is class distribution, and the more unimportant the cut point is. Yet, the difference of class distribution of adjacent two intervals which have the less number of classes is smaller and the corresponding value is smaller.

In this paper, we point out that using the importance of nodes determined by the distance, divided byfor extended Chi2 algorithm of reference [ 3 ] lacks theory basis and is not accurate. It is improbable to appear unreasonable factors. But, because the number of each group of adjacent intervals is different, it is unreasonable to merely take as a difference measure standard.

The theory analysis and the experiment results show that the presented algorithm is effective. The method uses test to determine whether the current point is merged or not. The significance test used in the algorithm requires training for selection of a confidence interval.

Journal of Applied Mathematics

Set the interval lower bound equal to the attribute value inclusive that belongs to this interval, and set its upper bound to the attribute value exclusive belonging to the next interval.

It should be merged.

But the method proposed in this paper is good. View at Google Scholar Z.