A Compact Evolutionary Interval-Valued Fuzzy Rule-Based Classification System for the Modeling and Prediction of Real-World Financial Applications With Imbalanced Data

Sanz, José Antonio; Bernardo, Dario; Herrera, Francisco; Bustince, Humberto; Hagras, Hani

doi:10.1109/tfuzz.2014.2336263

Cited by 150 publications

(49 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, and from a cognitive point of view, it comes out that, for these applications, the definition of parameters is not more complicated than the definition of the parameters for their fuzzy counterparts [125]. For instance: In classification problems [125], [126], [127], [128], [130]. The experimental results presented in [128] show that the approach using IVFSs (named IVTURS) improves the results of two state-of-the-art fuzzy classifiers like FARC-HD [2] and FURIA [77].…”

Section: -Measures Yielding Intervals From the Definition Ofmentioning

confidence: 99%

A Historical Account of Types of Fuzzy Sets and Their Relationships

Bustince

Barrenechea

Pagola

et al. 2016

IEEE Trans. Fuzzy Syst.

Self Cite

412

159

View full text Add to dashboard Cite

show abstract

Section: -Measures Yielding Intervals From the Definition Ofmentioning

confidence: 99%

A Historical Account of Types of Fuzzy Sets and Their Relationships

Bustince

Barrenechea

Pagola

et al. 2016

IEEE Trans. Fuzzy Syst.

Self Cite

412

159

View full text Add to dashboard Cite

show abstract

“…Focusing on classification problems composed of only two classes, the class having the largest number of examples is known as the majority class (it is also named negative class) whereas the remainder one is called minority class (or positive class). A wide number of real-world classification problems present the imbalanced issue 3,37,38,39 .…”

Section: Imbalanced Datasets Problemmentioning

confidence: 99%

“…Therefore, these techniques can be seen as classification systems 2 because their outcomes have two different values, namely, survive and die. Nowadays, the usage of intelligent systems has become a widely used solution to tackle classification problems 3,4,5 . Specifically, the standard intelligent system used by doctors to deal with the survival prediction problem is the logistic regression 8,9 , which obtains accurate results but it does not provide an explanation of its predictions.…”

Section: Introductionmentioning

confidence: 99%

A decision tree based approach with sampling techniques to predict the survival status of poly-trauma patients

Sanz¹,

Fernández²,

Bustince³

et al. 2017

IJCIS

Self Cite

View full text Add to dashboard Cite

Survival prediction of poly-trauma patients measure the quality of emergency services by comparing their predictions with the real outcomes. The aim of this paper is to tackle this problem applying C4.5 since it achieves accurate results and it provides interpretable models. Furthermore, we use sampling techniques because, among the 378 patients treated at the Hospital of Navarre, the number of survivals excels that of deaths. Logistic regressions are used in the comparison, since they are an standard in this domain.

show abstract

“…With the development of artificial intelligence technology, artificial neural network is commonly used to predict business failure [6]. Besides, there are other prediction methods, such as support vector machine [7], fuzzy sets theory [8] and so on. As undemanding hypothesis condition, these methods are more widely applied to BFP than previous prediction methods.…”

Section: Introductionmentioning

confidence: 99%

Business Failure Prediction using Fuzzy C-Means with Kernel Method

Wang¹,

Wu²,

Wu³

2017

dtssehs

View full text Add to dashboard Cite

Abstract:To propose the new prediction method of Kernel Fuzzy C-Means (KFCM) for business failure. Fuzzy C-Means (FCM) algorithm fails to deal with non-spherical clusters and incomplete data, while the kernel method can map the low-dimensional data into highdimensional feature space which is easier to be separated. Therefore, kernel method is integrated with the FCM to solve the problems of FCM. To fully reflect the performance of different kernel functions, KFCM respectively adopts three kernel functions which include Gaussian, Polynomial, Sigmoid kernel. The paper employs the financial data from Chinese quoted companies to predict the business failure. The prediction outcomes of three KFCMs are not only made a comparison to each other, but also compared with the standard FCM. It can show that KFCMs have better classification accuracy than FCM, but each of them has its advantages for normal and failing companies.

show abstract

A Compact Evolutionary Interval-Valued Fuzzy Rule-Based Classification System for the Modeling and Prediction of Real-World Financial Applications With Imbalanced Data

Cited by 150 publications

References 57 publications

A Historical Account of Types of Fuzzy Sets and Their Relationships

A Historical Account of Types of Fuzzy Sets and Their Relationships

A decision tree based approach with sampling techniques to predict the survival status of poly-trauma patients

Business Failure Prediction using Fuzzy C-Means with Kernel Method

Contact Info

Product

Resources

About