NSICA: Multi-objective imperialist competitive algorithm for feature selection in arrhythmia diagnosis

Ayar, Mehdi; Isazadeh, Ayaz; Gharehchopogh, Farhad Soleimanian; Seyedi, MirHojjat

doi:10.1016/j.compbiomed.2023.107025

Cited by 12 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This analysis revealed that RF achieved the highest accuracy (95.24%), precision (100%), sensitivity (89.47%), and specificity (100%) among all models. Furthermore, the proposed Random Forest model outperformed existing models referenced in the literature [8][9][10][11][12][13][14][15] across all performance metrics. To achieve this superior performance the authors employed standard data preprocessing and feature selection methods.…”

Section: Discussionmentioning

confidence: 84%

“…The proposed approach has attained a remarkable accuracy rate of 95.24%, surpassing the performance of the existing model, exhibiting noticeable (i.e., 2.05%, 9.26%, 7.97%, 46.43%, 20.56%, 0.85%, 1.86%, and 3.44%) enhancement than the reference research work. Furthermore, the proposed approach has exhibited superior precision (100), sensitivity (89.47%), and specificity (100%) as compared to [8][9][10][11][12][13][14][15]. The comparison of the proposed approach with existing methods has been summarised in figure 8.…”

Section: Benchmarkingmentioning

confidence: 99%

“…These two methods led to improved diagnosis accuracy for arrhythmias. Ayar et al (2023) [9] presented NSICA, a new multi-objective algorithm designed for optimal feature selection in arrhythmia classification, and demonstrated its use in both binary and multi-class tasks. Shahsavani et al (2023) [11] employed K-singular value decomposition (K-SVD), a sparse coding algorithm, for arrhythmia detection.…”

Section: Literature Surveymentioning

confidence: 99%

“…It portrays the model's predictions by displaying the number of true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN) [23]. By using these CM values, the authors analyzed the various performance metrics to determine the effectiveness, robustness, and efficiency of each model as displayed in equations (1) to (5) [9].…”

Section: Pre-processing Confusion Matrix and Performance Evaluation M...mentioning

confidence: 99%

See 3 more Smart Citations

A comparative study of heterogeneous machine learning algorithms for arrhythmia classification using feature selection technique and multi-dimensional datasets

Sharma,

Dhanka,

Kumar

et al. 2024

Eng. Res. Express

View full text Add to dashboard Cite

Arrhythmia, a common cardiovascular disorder, refers to the abnormal electrical activity within the heart, leading to irregular heart rhythms. This condition affects millions of people worldwide, with severe implications on cardiac function and overall health. Arrhythmias can strike anyone at any age which is a significant cause of morbidity and mortality on a global scale. About 80% of deaths related to heart disease are caused by ventricular arrhythmias. This research investigated the application of an optimized multi-objectives supervised Machine Learning (ML) models for early arrhythmia diagnosis. The authors evaluated the model's performance on the arrhythmia dataset from the UCI ML repository with varying train-test splits (70:30, 80:20, and 90:10). Standard preprocessing techniques such as handling missing values, formatting, balancing, and directory analysis were applied along with Pearson correlation for feature selection, all aimed at enhancing model performance. The proposed optimized RF model achieved impressive performance metrics, including accuracy (95.24%), precision (100%), sensitivity (89.47%), and specificity (100%). Furthermore, the study compared the proposed approach to existing models, demonstrating significant improvements across various performance measures.

show abstract

Section: Discussionmentioning

confidence: 84%

Section: Benchmarkingmentioning

confidence: 99%

Section: Literature Surveymentioning

confidence: 99%

Section: Pre-processing Confusion Matrix and Performance Evaluation M...mentioning

confidence: 99%

See 2 more Smart Citations

A comparative study of heterogeneous machine learning algorithms for arrhythmia classification using feature selection technique and multi-dimensional datasets

Sharma,

Dhanka,

Kumar

et al. 2024

Eng. Res. Express

View full text Add to dashboard Cite

show abstract

“…An e cient feature selection algorithm can improve the e ciency and performance of downstream training models. The feature selection method sorts features through algorithms, takes the upstream feature sorting results as input, and generates candidate feature subsets through feature subset search methods such as forward search [12][13][14][15][16]. This process has the problem of excessive dependence on feature sorting results and high dimensionality caused by mechanically generating feature subsets [17].…”

Section: Introductionmentioning

confidence: 99%

IIFS2.0: An Improved Incremental Feature Selection Method for Protein Sequence Processing Based on a Caching Strategy

Meng,

Pei,

et al. 2024

Preprint

View full text Add to dashboard Cite

The purpose of feature selection in protein sequence recognition problems is to select the optimal feature set and use it as training input for classifiers and discover key sequence features of specific proteins. In the feature selection process, relevant features associated with the target task will be retained, and irrelevant and redundant features will be removed. Therefore, in an ideal state, a feature combination with smaller feature dimensions and higher performance indicators is desired. This paper proposes an algorithm called IIFS2.0 based on the cache elimination strategy, which takes the local optimal combination of cached feature subsets as a breakthrough point. It searches for a new feature combination method through the cache elimination strategy to avoid the drawbacks of human factors and excessive reliance on feature sorting results. We validated and analyzed its effectiveness on the protein dataset, demonstrating that IIFS2.0 significantly reduces the dimensionality of feature combinations while also improving various evaluation indicators. In addition, we provide IIFS2.0 on http://112.124.26.17:8006/ for researchers to use.

show abstract

Multi-objective optimization enabling CFRP energy-efficient milling based on deep reinforcement learning

Zhang,

Yan

et al. 2024

Appl Intell

View full text Add to dashboard Cite

NSICA: Multi-objective imperialist competitive algorithm for feature selection in arrhythmia diagnosis

Cited by 12 publications

References 23 publications

A comparative study of heterogeneous machine learning algorithms for arrhythmia classification using feature selection technique and multi-dimensional datasets

A comparative study of heterogeneous machine learning algorithms for arrhythmia classification using feature selection technique and multi-dimensional datasets

IIFS2.0: An Improved Incremental Feature Selection Method for Protein Sequence Processing Based on a Caching Strategy

Multi-objective optimization enabling CFRP energy-efficient milling based on deep reinforcement learning

Contact Info

Product

Resources

About