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: Jugrel Voodookinos
Country: Montenegro
Language: English (Spanish)
Genre: Automotive
Published (Last): 13 April 2014
Pages: 385
PDF File Size: 17.35 Mb
ePub File Size: 7.84 Mb
ISBN: 435-3-22900-361-1
Downloads: 90250
Price: Free* [*Free Regsitration Required]
Uploader: Arashizahn

It tests the following hypothesis: Approximate reasoning is an important research content of artificial intelligence domain [ 14 — 17 ]. By continuing to use this website, you agree to their use. Rectified Chi2 algorithm proposed in this paper controls merger extent and information loss in the discretization process with. 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.

A good similarity measure should have the following characteristic: In brief, interval similarity definition not only can inherit the logical aspects of statistic but also can resolve the problems about algorithms of the correlation of Chi2 algorithm, realizing equality. It is improbable to appear unreasonable factors. Abstract Discretization algorithm for real value attributes is of very important uses in many areas such as intelligence and machine learning.

This site uses cookies. The algorithm SIM is as shown in Algorithm 2. Three Classes are available which are Iris-setosa, Iris-versicolorIris-virginica. The ChiMerge algorithm introduced by Kerber in is a supervised global discretization method [ 11 ]. The number of samples of two intervals is the same. At the same time, based on the study of applied meaning of statistic, the drawback of the algorithm is analyzed.

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

Below is an easy way of doing this:. B Create a frequency table containing one row for each distinct attribute value and one column for each class. Then generate a set of distinct intervals.


Email Subscription Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Attribtues, the degree of freedom should be determined by the number of decision classes of each two adjacent intervals. Below are my final results compared to the results on http: Based on the analysis to the drawback of the correlation of Chi2 algorithm, we propose the similarity function as follows. Moreover, degree of freedom of adjacent two disccretization with the greater number of classes is bigger. However, from the experiments we can see that SIM algorithm does not outperform extended Chi2 algorithm and Boolean discretization algorithm for all datasets.

In particular promotion scope of Glass, Wine, and Machine datasets is very big.

ChiMerge discretization algorithm

Thus, if extended Chi2 discretization algorithm was used, it is not accurate and unreasonable to merge first adjacent two intervals which have the maximal difference value. In formula 3under certain situations is not very accurate: Comparison of distribution with different degrees of freedom. This time, merged standard of extended Chi2 algorithm is possibly more accurate in computation.

In [ 3 ], the authors pointed out that the method of calculating the freedom degrees in the modified Chi2 algorithm is not accurate and proposed the extended Chi2 algorithm, which replaced with. The formula for computing the value is where: The average predictive accuracy, the average numbers of nodes fiscretization decision tree, and the average numbers of rules extracted are computed and compared by different algorithms see Table 3.

Under the comparison for two methods, the difference of recognition and forecast effect of Auto and Iris datasets each of them has off classes is small. For the newest extended Chi2 algorithm, it is very possible to have such two groups of adjacent intervals: Finally, we are ready to implement the Chi-Merge algorithm.


Ionosphere and Wine datasets. Journal of Applied Mathematics. But in fact, it is possibly unreasonable that they are first merged.

Reference [ 7 ] is an algorithm for discretization of real value attributes based on decision table and information entropy, which belongs to a heuristic and local algorithm that seeks the best results.

Considering any adjacent two intervals andcan express the difference degree between adjacent two intervals. So, when value is equal to 0, using difference as the standard of interval merging is inaccurate. Model type aftributes C-SVC. View at Google Scholar. In machine learning and data mining, many algorithms have already been developed according to processing discrete data. In some domain such as picture matching, information retrieval, computer vision, image fusion, remote sensing, and weather forecast, similarity measure has the extremely vital significance [ 1319 — 22 ].

Post was not sent – check nmeric email addresses! Net 30 Algorithms 17 Databases 4 Errors!

Once you run the algorithm, you can compare the obtained intervals and split points with the ones found on the internet. Check these ChiMerge powerpoint slides that visualizes the above algorithm. The significance test used in the algorithm requires training for selection of a confidence interval.

Even if degree of freedom in is bigger thanbut because the difference of degree of freedom between and is very small, it is possible that the difference of is bigger than the difference of. Regarding such situation, the method proposed in this paper has superiority very well e.

View at Google Scholar S.