Enhancing Linear Independent Component Analysis: Comparison of Various Metaheuristic Methods

Abbas, Nidaa A.; Salman, Hussein

doi:10.37917/ijeee.16.1.14

Cited by 5 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When these traditional methods fail, ICA is a far more effective method that can uncover the underlying causes or sources. The FastICA algorithm was used in this study [ 43 , 44 ].…”

Section: Methodsmentioning

confidence: 99%

Prediction of diabetes disease using an ensemble of machine learning multi-classifier models

Abnoosian,

Farnoosh,

Behzadi

2023

BMC Bioinformatics

View full text Add to dashboard Cite

Background and objective Diabetes is a life-threatening chronic disease with a growing global prevalence, necessitating early diagnosis and treatment to prevent severe complications. Machine learning has emerged as a promising approach for diabetes diagnosis, but challenges such as limited labeled data, frequent missing values, and dataset imbalance hinder the development of accurate prediction models. Therefore, a novel framework is required to address these challenges and improve performance. Methods In this study, we propose an innovative pipeline-based multi-classification framework to predict diabetes in three classes: diabetic, non-diabetic, and prediabetes, using the imbalanced Iraqi Patient Dataset of Diabetes. Our framework incorporates various pre-processing techniques, including duplicate sample removal, attribute conversion, missing value imputation, data normalization and standardization, feature selection, and k-fold cross-validation. Furthermore, we implement multiple machine learning models, such as k-NN, SVM, DT, RF, AdaBoost, and GNB, and introduce a weighted ensemble approach based on the Area Under the Receiver Operating Characteristic Curve (AUC) to address dataset imbalance. Performance optimization is achieved through grid search and Bayesian optimization for hyper-parameter tuning. Results Our proposed model outperforms other machine learning models, including k-NN, SVM, DT, RF, AdaBoost, and GNB, in predicting diabetes. The model achieves high average accuracy, precision, recall, F1-score, and AUC values of 0.9887, 0.9861, 0.9792, 0.9851, and 0.999, respectively. Conclusion Our pipeline-based multi-classification framework demonstrates promising results in accurately predicting diabetes using an imbalanced dataset of Iraqi diabetic patients. The proposed framework addresses the challenges associated with limited labeled data, missing values, and dataset imbalance, leading to improved prediction performance. This study highlights the potential of machine learning techniques in diabetes diagnosis and management, and the proposed framework can serve as a valuable tool for accurate prediction and improved patient care. Further research can build upon our work to refine and optimize the framework and explore its applicability in diverse datasets and populations.

show abstract

“…When these traditional methods fail, ICA is a far more effective method that can uncover the underlying causes or sources. The FastICA algorithm was used in this study [ 43 , 44 ].…”

Section: Methodsmentioning

confidence: 99%

Prediction of diabetes disease using an ensemble of machine learning multi-classifier models

Abnoosian,

Farnoosh,

Behzadi

2023

BMC Bioinformatics

View full text Add to dashboard Cite

show abstract

“…Where represents the -th cumulant, is an expectation operation, and is data vector of the signals [1], [7], [14], [15].…”

Section: Independent Component Analysis (Ica)mentioning

confidence: 99%

“…Abbas and Salman [15] introduced some methods to enhance the performance of the linear ICA dependening on the quantum particle swarm optimization (QPSO) and the gloworm swarm optimization (GSO) with three objective functions are Entropy, Negentropy, and Mutual Information. So, the author proposed new Nonlinear ICA method depends on some nonlinear methods.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Bigradient neural network-based quantum particle swarm optimization for blind source separation

Salman

Al-Qurabat

Finjan

2021

IJ-AI

View full text Add to dashboard Cite

An independent component analysis (ICA) is one of the solutions of a blind source separation problem. ICA is a statistical approach that depends on the statistical properties of the mixed signals. The purpose of the ICA method is to demix the mixed source signals (observation signals) and rcovering those signals. The abbreviation of the problem is that the ICA needs for optimizing by using one of the optimization approaches as swarm intelligent, neural neworks, and genetic algorithms. This paper presents a hybrid method to optimize the ICA method by using the quantum particle swarm optimization method (QPSO) to optimize the Bigradient neural network method that applies to separate mixed signals and recover sources signals. The results of an implement this work prove that this method gave good results comparing with other methods such as the Bigradient neural network and the QPSO method, based on several evaluation measures as signal-to-noise ratio, signal-to-distortion ratio, absolute value correlation coefficient, and the computation time.

show abstract

“…The second part has effects on the algorithmic features. Therefore, the ICA is stronger when it includes a strong objective function, which means a fast and simple computation [11], [13].…”

Section: Introductionmentioning

confidence: 99%

Solve Cocktail Party Problem Based on Hybrid Method

M. Salman,

Kadhum M. Al-Qurabat,

Riyadh Finjan

2024

IJCDS

View full text Add to dashboard Cite

One of the most intractable issues in contemporary digital signal processing, particularly with regards to blind source separation methods, is known as the cocktail party dilemma. This problem suppose there are many sensors record many signals at same time to produce many mixed signals. To solve this problem, one of an important methods used for this purpose is an Independent Component Analysis method. This method abbreviates in how separate mixed signals without any pre-knowledge about the mixing signals?. It treats on the statistical features of a mixing signals. This work introduces a novel method to solve the cocktail party problem, by using hybrid method from the Quantum Particle Swarm Optimization method and the Bell-Sejnowski neural method to enhance the performance of the Independent Component Analysis. In addition, the proposed method uses the Negentropy function to be the objective function of the optimization process. The proposed algorithm has been implemented on two cases of three really signals, with 8-KHz frequencies. The results of the separating process measured in two directions: firstly by comparing the results with other methods as Particle Swarm Optimization and the Quantum Particle Swarm Optimization, where the results appear that the proposed method appears very high results than other methods. Secondly, by using standard metrics as Absolute Value Correlation Coefficient, Signal to Distortion Ratio, and Signal to Noise Ratio.

show abstract

Enhancing Linear Independent Component Analysis: Comparison of Various Metaheuristic Methods

Cited by 5 publications

References 0 publications

Prediction of diabetes disease using an ensemble of machine learning multi-classifier models

Prediction of diabetes disease using an ensemble of machine learning multi-classifier models

Bigradient neural network-based quantum particle swarm optimization for blind source separation

Solve Cocktail Party Problem Based on Hybrid Method

Contact Info

Product

Resources

About