Fractional cuckoo search invasive weed optimized neural network for data classification
Authors:
- 14 November 2023
- journal article
- research article
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 36 (6)
Abstract
Summary: The progression of organizing and classifying data components based on pre‐established criteria is known as data classification. In data classification problems, machine learning methods are frequently used. Although the ability of an algorithm to learn is essential, the recognition quality of algorithms is undoubtedly a key indicator. However, the error made in the beginning is significant, and if it is not rectified then, it causes chaos in data classification. In this research, a fractional hybrid optimization is created with a feed‐forward neural network (FFNN) for the data classification. The pre‐processing of the medical data involves the imputation of missing values and the log transformation. In the feature selection process, the chord distance and the Jaro‐Winkler distance are both employed to choose the important features. The FFNN classifier is used to classify the data, which is trained using the hybrid optimization method, namely the developed Fractional Cuckoo search Invasive Weed Optimization algorithm (FCSIWO). The fractional concept, Invasive Weed Optimization (IWO) algorithm, and Cuckoo search (CS) algorithm are combined to create the FCSIWO. When compared to existing data classification models, the developed technique achieved an optimal sensitivity of 0.9463, an optimal accuracy of 0.9440, and an optimal specificity of 0.9422.Keywords
This publication has 21 references indexed in Scilit:
- Improving Extreme Learning Machine by Competitive Swarm Optimization and its application for medical diagnosis problemsExpert Systems with Applications, 2018
- Classification of Imbalanced Data by Oversampling in Kernel Space of Support Vector MachinesIEEE Transactions on Neural Networks and Learning Systems, 2017
- Data classification with deep learning using Tensorflow2017
- Weighted extreme learning machine for imbalance learningNeurocomputing, 2013
- A Review on Ensembles for the Class Imbalance Problem: Bagging-, Boosting-, and Hybrid-Based ApproachesIEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 2012
- Hellinger distance decision trees are robust and skew-insensitiveData Mining and Knowledge Discovery, 2011
- Invasive Weed Optimization and its Features in ElectromagneticsIEEE Transactions on Antennas and Propagation, 2010
- Engineering optimisation by cuckoo searchInternational Journal of Mathematical Modelling and Numerical Optimisation, 2010
- Exploratory Undersampling for Class-Imbalance LearningIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2008
- SVMs Modeling for Highly Imbalanced ClassificationIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2008