Fast Correlation Based Filter (FCBF) with a different search strategy

Senliol, Baris; Gülgezen, Gökhan; Yu, Lei; Çataltepe, Zehra

doi:10.1109/iscis.2008.4717949

Cited by 100 publications

(52 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The FCBF algorithm, a filtering method, can identify the value of a feature in a classification by calculating the correlation values between features and class labels or correlation values between 2 features. By doing so, it does not need the results of classification algorithms [34]. This characteristic provides a great advantage over spiral feature selection algorithms in terms of speed.…”

Section: Redundancy Analysismentioning

confidence: 98%

“…The algorithm is an efficient feature selection algorithm based on the relevance among features and redundancy values ( Figure 5). The FCBF algorithm is a multivariate feature selection method starting with a full set of features, using symmetrical uncertainty (SU) to calculate the dependences of features and arriving at the best subset, using a backward selection technique with a sequential search strategy [34].…”

Section: Fcbf Selection Algorithmmentioning

confidence: 99%

“…Linear correlation computations may provide inaccurate results in situations where the correlations among the features or the correlations between a feature and a class label are not linear [34]. The SU is defined as:…”

Section: Data Set Relevance Analysismentioning

confidence: 99%

“…Figure 6 displays the list of SU i,c values obtained in this study. Since the use of threshold values bigger than 0 may cause the elimination of some features in the first phase, it is actually effective in accelerating the algorithm [34]. However, since this study is not related to the analysis of the speed but to the performance, the threshold value of 0 is accepted.…”

Section: Data Set Relevance Analysismentioning

confidence: 99%

See 3 more Smart Citations

Novel approaches for automated epileptic diagnosis using FCBF selection and classification algorithms

Şen¹,

Peker²

2013

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

This paper presents a new application for automated epileptic detection using the fast correlation-based feature (FCBF) selection and classification algorithms. This study consists of 3 stages: feature extraction, feature selection from electroencephalography (EEG) signals, and the classification of these signals. In the feature extraction phase, 16 attribute algorithms are used in 5 categories, and 36 feature parameters are obtained from these algorithms.In the feature selection phase, the FCBF algorithm is chosen to select a set of attributes that best represent the EEG signals. The resulting attributes are used as input parameters for the classification algorithms. In the classification phase, the problem is classified with 6 different classification algorithms. The results obtained with the different classification algorithms are provided in order to compare the calculation times and the accuracy rates. The evolution of the proposed system is conducted using k-fold cross-validation, classification accuracy, sensitivity and specificity values, and a confusion matrix. The proposed approach enables 100% classification accuracy with the use of the multilayer perceptron neural network and naive Bayes algorithm. The stated results show that the proposed method is capable of designing a new intelligent assistance diagnostic system.

show abstract

Section: Redundancy Analysismentioning

confidence: 98%

Section: Fcbf Selection Algorithmmentioning

confidence: 99%

Section: Data Set Relevance Analysismentioning

confidence: 99%

Section: Data Set Relevance Analysismentioning

confidence: 99%

See 2 more Smart Citations

Novel approaches for automated epileptic diagnosis using FCBF selection and classification algorithms

Şen¹,

Peker²

2013

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…Thyroid function diagnosis is an important classification issue. Proper interpretation of the thyroid data, besides clinical examination and complementary investigation, is an important problem in the diagnosis of thyroid disease [1,2,4]. …”

Section: Introductionmentioning

confidence: 99%