A discretization algorithm based on Class-Attribute Contingency Coefficient

Tsai, Chao-Ming; Lee, Chien-I; Yang, Wei‐Pang

doi:10.1016/j.ins.2007.09.004

Cited by 184 publications

(104 citation statements)

References 28 publications

(35 reference statements)

Supporting

Mentioning

100

Contrasting

Unclassified

Order By: Relevance

“…As can be seen, the discretization methods chosen would affect the final classification result. But to the best of our knowledge [16] [17] and also testing results, so far no particular discretization method is clearly superior to the others for our data. Thus, the discretization method applied here is the standard one embedded in WEKA, which is minimum description length (MDL) [18] [19].…”

Section: Experiments and Resultsmentioning

confidence: 64%

Feature Selection Using Correlation and Reliability Based Scoring Metric for Video Semantic Detection

Zhu

Lin

Shyu

et al. 2010

2010 IEEE Fourth International Conference on Semantic Computing

View full text Add to dashboard Cite

Abstract-Content-based multimedia retrieval faces many challenges such as semantic gap, imbalanced data, and varied qualities of the media. Feature selection as a component of the retrieval process plays an important role. The aim of feature selection is to identify a subset of features by removing irrelevant or redundant features. An effective subset of features can not only improve model performance and reduce computational complexity, but also enhance semantic interpretability. To achieve these objectives, in this paper, a novel metric that integrates the correlation and reliability information between each feature and each class obtained from Multiple Correspondence Analysis (MCA) is proposed to score the features for feature selection. Based on these scores, a ranked list of features can be generated and different selection criteria can be adopted to select a subset of features. To evaluate the proposed framework, four other wellknown feature selection methods, namely information gain, chisquare measure, correlation-based feature selection, and relief are compared with the proposed method over five popular classifiers using the benchmark data from TRECVID 2009 highlevel feature extraction task. The results show that the proposed method outperforms the other methods in terms of classification accuracy, the size of feature subspace, and the ability to capture the semantic information.

show abstract

Section: Experiments and Resultsmentioning

confidence: 64%

Feature Selection Using Correlation and Reliability Based Scoring Metric for Video Semantic Detection

Zhu

Lin

Shyu

et al. 2010

2010 IEEE Fourth International Conference on Semantic Computing

View full text Add to dashboard Cite

show abstract

“…These datasets have been frequently used as a benchmark to compare the performance of classification methods and consist of a mixture of numeric, real and categorical attributes. Numeric features are prediscretized by the method demonstrated in [23], which begins by sorting a dataset and selecting only duplicate values for the cutting point bin. After this step, the number of discrete values to represent each bin is found.…”

Section: Experimental Evaluationsmentioning

confidence: 99%

Improving Floating Search Feature Selection using Genetic Algorithm

Homsapaya

Sornil

2017

J. ICT Res. Appl.

View full text Add to dashboard Cite

Abstract. Classification, a process for predicting the class of a given input data, is one of the most fundamental tasks in data mining. Classification performance is negatively affected by noisy data and therefore selecting features relevant to the problem is a critical step in classification, especially when applied to large datasets. In this article, a novel filter-based floating search technique for feature selection to select an optimal set of features for classification purposes is proposed. A genetic algorithm is employed to improve the quality of the features selected by the floating search method in each iteration. A criterion function is applied to select relevant and high-quality features that can improve classification accuracy. The proposed method was evaluated using 20 standard machine learning datasets of various size and complexity. The results show that the proposed method is effective in general across different classifiers and performs well in comparison with recently reported techniques. In addition, the application of the proposed method with support vector machine provides the best performance among the classifiers studied and outperformed previous researches with the majority of data sets.

show abstract

“…So our system involves discretizing the continuous attributes based on the classification pre-determined class target. For discretization, we will use the algorithm from [11] which is implemented by Guangdi Li.…”

Section: Pre-processing Phasementioning

confidence: 99%

A Classification Technique using Associative Classification

Shekhawat¹,

Dhande²

2011

IJCA

View full text Add to dashboard Cite

Classification and association rule mining are two basic tasks of Data Mining. Classification rule mining is used to discover a small set of rules in the database to form an accurate classifier. Association rules mining has been used to reveal all interesting relationships in a potentially large database. An Apriori approach, which was used to generate the association rules from frequent patterns, turn out to generate a huge time-intensive query called as iceberg query. Various researches have been done under the Apriori-like approach to improve performance of the frequent pattern mining tasks but the results were not as much as expected due to many scans on the dataset. This project aims to propose a flexible way of mining frequent patterns by extending the idea of the Associative Classification methods. For better performance, the Neural Network Association Classification system is proposed here to be one of the approaches for building accurate and efficient classifiers. In this paper, the Neural Network Association Classification system is used in order to improve its accuracy. The structure of the network reflects the knowledge uncovered in the previous discovery phase. The trained network is then used to classify unseen data. The performance of the Neural Network Associative Classification system is compared with the previous Classification Based Association on four datasets from UCI machine learning repository.

show abstract

A discretization algorithm based on Class-Attribute Contingency Coefficient

Cited by 184 publications

References 28 publications

Feature Selection Using Correlation and Reliability Based Scoring Metric for Video Semantic Detection

Feature Selection Using Correlation and Reliability Based Scoring Metric for Video Semantic Detection

Improving Floating Search Feature Selection using Genetic Algorithm

A Classification Technique using Associative Classification

Contact Info

Product

Resources

About