A Comparison of Six Approaches to Discretization—A Rough Set Perspective

Blajdo, Piotr; Grzymała-Busse, Jerzy W.; Hippe, Zdzisław S.; Knap, Maksymilian; Mroczek, Teresa; Piątek, Łukasz

doi:10.1007/978-3-540-79721-0_10

Cited by 15 publications

(9 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These new methods of discretization are called the globalized version of equal interval width and the globalized version of equal frequency per interval. As follows from [2], both methods are quite successful.…”

Section: Equal Interval Width and Equal Frequency Per Intervalmentioning

confidence: 93%

See 1 more Smart Citation

A Comparison of Four Approaches to Discretization Based on Entropy

Grzymała-Busse

Mroczek

2016

Entropy

Self Cite

View full text Add to dashboard Cite

Abstract:We compare four discretization methods, all based on entropy: the original C4.5 approach to discretization, two globalized methods, known as equal interval width and equal frequency per interval, and a relatively new method for discretization called multiple scanning using the C4.5 decision tree generation system. The main objective of our research is to compare the quality of these four methods using two criteria: an error rate evaluated by ten-fold cross-validation and the size of the decision tree generated by C4.5. Our results show that multiple scanning is the best discretization method in terms of the error rate and that decision trees generated from datasets discretized by multiple scanning are simpler than decision trees generated directly by C4.5 or generated from datasets discretized by both globalized discretization methods.

show abstract

Section: Equal Interval Width and Equal Frequency Per Intervalmentioning

confidence: 93%

“…Discretization based on the conditional entropy of the concept given the attribute is considered to be one of the most successful discretization techniques [2,5,6,9,11,12,14,15,19,24,26,27]. Let a be an attribute and q be a cut point that splits the set S into two subsets, S 1 and S 2 .…”

Section: Discretizationmentioning

confidence: 99%

A Comparison of Four Approaches to Discretization Based on Entropy

Grzymała-Busse

Mroczek

2016

Entropy

Self Cite

View full text Add to dashboard Cite

show abstract

“…(Please refer to [38] and [45] for further explanations on discretization technique. It is essential to choose an appropriate discretization method since performance of discretization methods differ significantly [46]. The experiments were conducted in order to choose the most significant discretization method for our data.…”

Section: Rules Analysis Of the Integrated Felder Silverman Learning Smentioning

confidence: 99%

Granular Mining of Student’s Learning Behavior in Learning Management System Using Rough Set Technique

Ahmad

Shamsuddin

Abraham

2010

Studies in Computational Intelligence

View full text Add to dashboard Cite

Abstract. Pattern multiplicity of user interaction in learning management system can be intelligently examined to diagnose students' learning style. Such patterns include the way the user navigate, the choice of the link provided in the system, the preferences of type of learning material, and the usage of the tool provided in the system. In this study, we propose mapping development of student characteristics into Integrated Felder Silverman (IFS) learning style dimensions. Four learning dimensions in Felder Silverman model are incorporated to map the student characteristics into sixteen learning styles. Subsequently, by employing rough set technique, twenty attributes have been selected for mapping principle. However, rough set generates a large number of rules that might have redundancy and irrelevant. Hence, in this study, we assess and mining the most significant IFS rules for user behavior by filtering these irrelevant rules. The assessments of the rules are executed by evaluating the rules support, the rules length and the accuracy. The irrelevant rules are further filtered by measuring different rules support, rules length and rules accuracy. It is scrutinized that the rules with the length in between [4,8], and the rules support in the range of [6,43] succumb the highest accuracy with 96.62%.

show abstract

“…Singh and Minz proposed a discretization approach implemented on grouping and coarse set conjecture [4]. Blajdo et al compared the results of six promising discretization approaches from the standpoint of rough sets [5]. Tian et al proposed a core-generating discretization method, which was used as the pre-processor of coarse setbased article assortment [6].…”

Section: Introductionmentioning

confidence: 99%

An Improved SVM Classifier for Discretization of Attributes using K-Means Clustering

Dixit¹

2017

IJCA

View full text Add to dashboard Cite

Here in this broadside a novel approach for the Discretization of Nonstop Characteristics for the Classification of various datasets is proposed. The Planned Procedure implemented here works in Two Phases, in the first stage K-means Clustering is applied on the dataset to cluster the data on the basis of classes available in the dataset and second is to classify the Clustered Data using Support Vector Machine Classifier. The various Untried results achieved on different datasets proves that the planned procedure provides less mean number of cuts and reduced mean discretization time and also provides higher accuracy with better Scalability.

show abstract

A Comparison of Six Approaches to Discretization—A Rough Set Perspective

Cited by 15 publications

References 5 publications

A Comparison of Four Approaches to Discretization Based on Entropy

A Comparison of Four Approaches to Discretization Based on Entropy

Granular Mining of Student’s Learning Behavior in Learning Management System Using Rough Set Technique

An Improved SVM Classifier for Discretization of Attributes using K-Means Clustering

Contact Info

Product

Resources

About