Ideal Combination Feature Selection Model for Classification Problem based on Bio-Inspired Approach

机译：基于生物启发方法的分类问题理想组合特征选择模型

获取原文

页面导航

摘要
著录项
相似文献
相关主题

摘要

Feature selection or attribute reduction is a crucial process to achieve optimal data reduction for classification task. However, most of the feature selection methods that were introduced work individually that sometimes caused less optimal feature being selected, subsequently degrading the consistency of the classification accuracy rate. The aim of this paper is to exploit the capability of bio-inspired search algorithms, together with wrapper and filtered methods in generating optimal set of features. The important step is to idealize the combined feature selection models by finding the best combination of search method and feature selection algorithms. The next step is to define an optimized feature set for classification task. Performance metrics are analyzed based on classification accuracy and the number of selected features. Experiments were conducted on nine (9) benchmark datasets with various sizes, categorized as small, medium and large dataset. Experimental results revealed that the ideal combination is a feature selection model with the implementation of bio-inspired search algorithm that consistently obtains the optimal solution (i.e. less number of features with higher classification accuracy) on the selected dataset. Such a finding indicates that the exploitation of bio-inspired algorithms with ideal combination of wrapper/filtered method can contribute in finding the optimal features to be used in data mining model construction.

机译：特征选择或属性缩减是实现分类任务的最佳数据减少的重要过程。然而，大多数特征选择方法都是单独推出的，有时会引起更少的最佳特征，随后降低分类精度率的一致性。本文的目的是利用生物启发搜索算法的能力以及包装器和过滤的方法生成最佳特征。重要的步骤是通过找到搜索方法和特征选择算法的最佳组合来理解组合特征选择模型。下一步是为分类任务定义设置的优化功能。根据分类准确性和所选功能的数量分析性能指标。在九（9）个基准数据集中进行实验，具有各种尺寸，分为小，中等和大型数据集。实验结果表明，理想的组合是具有生物启发搜索算法的特征选择模型，该搜索算法一致地在所选数据集上始终获得最佳解决方案（即，具有较高分类精度的特征数量）。这种发现表明，具有理想组合的包装/过滤方法的生物启发算法的利用可以有助于找到数据挖掘模型结构中的最佳特征。

著录项

来源
《International conference on omputational science and technology》|2020年|xv 734 p.|共9页
会议地点
作者
Mohammad Aizat Basir; Mohamed Saifullah Hussin; Yuhanis Yusof;
展开▼
作者单位

展开▼
会议组织
原文格式 PDF
正文语种
中图分类计算技术、计算机技术;
关键词
Feature Selection; Bio-Inspired; Classification;

机译：特征选择;生物启发;分类;

相似文献

外文文献
中文文献
专利

1. Modelling of F3I based feature selection approach for PCOS classification and prediction [J] . Maheswari K., Baranidharan T., Karthik S., Journal of ambient intelligence and humanized computing . 2021,第1期

机译：基于F3I的PCOS分类和预测特征选择方法的建模
2. Feature Selection for Object-Based Classification of High-Resolution Remote Sensing Images Based on the Combination of a Genetic Algorithm and Tabu Search [J] . Lei Shi, Youchuan Wan, Xianjun Gao, Computational intelligence and neuroscience . 2018,第3期

机译：遗传算法与禁忌搜索相结合的高分辨率遥感影像基于对象分类的特征选择
3. Feature Selection for Object-Based Classification of High-Resolution Remote Sensing Images Based on the Combination of a Genetic Algorithm and Tabu Search [J] . Lei Shi, Youchuan Wan, Xianjun Gao, Computational intelligence and neuroscience . 2018,第Pta1期

机译：基于对象的基于对象的分类的特征选择，高分辨率遥感图像的基于遗传算法和禁忌搜索的组合
4. Ideal Combination Feature Selection Model for Classification Problem based on Bio-Inspired Approach [C] . Mohammad Aizat Basir, Mohamed Saifullah Hussin, Yuhanis Yusof International conference on omputational science and technology . 2020

机译：基于生物启发方法的分类问题理想组合特征选择模型
5. Statistical model-based methods for observation selection in wireless sensor networks and for feature selection in classification. [D] . Qi, Qi. 2012

机译：基于统计模型的方法用于无线传感器网络中的观察选择和分类中的特征选择。
6. Feature Selection for Object-Based Classification of High-Resolution Remote Sensing Images Based on the Combination of a Genetic Algorithm and Tabu Search [O] . Lei Shi, Youchuan Wan, Xianjun Gao, 2018

机译：遗传算法与禁忌搜索相结合的高分辨率遥感影像基于对象分类的特征选择
7. The thermodynamic changes that occur upon mixing five models of formamide and three models of water, including the miscibility of these model combinations itself, is studied by performing Monte Carlo computer simulations using an appropriately chosen thermodynamic cycle and the method of thermodynamic integration. The results show that the mixing of these two components is close to the ideal mixing, as both the energy and entropy of mixing turn out to be rather close to the ideal term in the entire composition range. Concerning the energy of mixing, the OPLS/AA-mod model of formamide behaves in a qualitatively different way than the other models considered. Thus, this model results in negative, while the other ones in positive energy of mixing values in combination with all three water models considered. Experimental data supports this latter behavior. Although the Helmholtz free energy of mixing always turns out to be negative in the entire composition range, the majority of the model combinations tested either show limited miscibility, or, at least, approach the miscibility limit very closely in certain compositions. Concerning both the miscibility and the energy of mixing of these model combinations, we recommend the use of the combination of the CHARMM formamide and TIP4P water models in simulations of water-formamide mixtures. [O] . Kiss, Bálint, Fábián, Balázs, Idrissi, Abdenacer, 2017

机译：通过使用适当选择的热力学循环和热力学积分方法进行蒙特卡罗计算机模拟，研究了将五种甲酰胺模型和三种水模型混合在一起时发生的热力学变化，包括这些模型组合本身的可混溶性。结果表明这两种组分的混合接近于理想的混合，因为混合的能量和熵在整个组成范围内都非常接近理想的项。关于混合的能量，甲酰胺的OPLS / AA-mod模型与其他模型相比，在质量上有不同的表现。因此，该模型得出的结果是负的，而其他模型则综合考虑了所有三个水模型的结果的正能量。实验数据支持后一种行为。尽管混合的亥姆霍兹自由能在整个组成范围内始终为负，但大多数测试模型组合显示出有限的混溶性，或至少非常接近某些组合物的混溶性极限。关于这些模型组合的可混溶性和混合能量，我们建议在水-甲酰胺混合物的模拟中使用CHARMM甲酰胺和TIP4P水模型的组合。

Ideal Combination Feature Selection Model for Classification Problem based on Bio-Inspired Approach

摘要

著录项

相似文献

相关主题

期刊订阅