Online ISSN: 2515-8260

Keywords : Particle swarm optimization


Hybrid of PSO-GSA based Clustering Approach for Predicting Structural Class Prediction using Random Forest Method

Sarneet Kaur; Ashok Sharma; Parveen Singh

European Journal of Molecular & Clinical Medicine, 2020, Volume 7, Issue 10, Pages 17-32

Protein can be classified in different classes. A lot of research is being performed for analyzing the structure and classes of protein. There are many problems associated with protein structure. Some of them are folding problem and protein structure prediction (PSP) etc. PSP is the most considerable open problem in field of biology. In the present work different algorithms like particle swarm optimization (PSO), gravitational search algorithm (GSA) and K-Mean clustering algorithms are used to classify different structures of protein. A random forest (RF) classifier is proposed for analyzing and comparing different protein classes in terms of other conventionally available algorithms in terms of various performance parameters like accuracy, recall, precision and specificity. The proposed classifier proved better than other classifiers in terms of accuracy and can be helpful in predicting the protein structures. A hybrid PSO-GSA algorithm is also proposed which provided improved performance as compared to single algorithms and can be utilized for analysis of protein structure.

Protein Structural Classes Prediction Based On Convolutional Neural Network Classifier with Feature Selection of Hybrid PSO-FA Optimization Approach

Sarneet Kaur; Ashok Sharma; Parveen Singh

European Journal of Molecular & Clinical Medicine, 2020, Volume 7, Issue 10, Pages 252-265

Protein can be classified in different classes like A (All α), B (All β), C (α+β), and D (α/β). A lot of work has been performed for analyzing the Sub-cellular localization of protein structure. The visualization of protein folding into compact conformation is evaluated. In the present work different algorithms like particle swarm optimization (PSO), Firefly algorithm (FFA) and K-Mean clustering algorithms are used to classify different structures of protein. A Conventional neural network (CNN) classifier is utilized for analyzing and comparing different protein classes in terms of SVM classifier available conventionally in terms of various performance parameters. Near 100 % accuracy, sensitivity, specificity, and MCC values are obtained for class A & class B protein structures. However, somewhat lower values of these parameters are obtained for class C and class D protein structures. CNN classifier proved better than SVM classifier and can be helpful in predicting the protein structures. A hybrid PSO-FFA algorithm is used to extract the features for different classes of protein. Structures of four classes of protein are evaluated in terms of scoring spaces and fitnessvalues.

ADAPTIVE DIMENSIONAL PARTICLE SWARM OPTIMIZATION BASED HYPER BASIS FUNCTION NEURAL NETWORK CLASSIFICATION FOR SOFTWARE FAILURE CAUSE PREDICTION

Mr.P. SARAVANAN1; Dr.V. SANGEETHA

European Journal of Molecular & Clinical Medicine, 2020, Volume 7, Issue 9, Pages 956-969

In software engineering, Detection of software root cause failure is a considerable issue to be resolved for increasing the reliability. Few research works are introduced for predicting the software failure causes with help of diverse classification algorithm. However, False Positive Rate (FPR) of failure detection process was higher. Therefore, a novel software failure cause prediction model called Adaptive Dimensional Particle Swarm Optimization Based Hyper Basis Function Neural Network (ADPSO-HBFNN) Model is proposed to increase the software reliability through predicting the root cause of software failure at an early stage. ADPSO-HBFNN Model initially gets number of event log files as input. Next, ADPSO-HBFNN Model applies Hyper Basis Function Neural Networks (HBFNNs) for discovering the software fault root cause by means of classifying the event log files. Subsequently, ADPSO-HBFNN Model applies Adaptive Dimensional Search Based Particle Swarm Optimization (ADS-PSO) algorithm where it considers the cost sensitive factor such as expected cost of software failure misclassification. The ADS-PSO algorithm lessen mean square error (MSE) during the learning process by optimizing the weights of network. From that, ADPSO-HBFNN Model correctly find outs the root cause of software failure with higher accuracy. Simulation outcome of ADPSO-HBFNN Model increase the accuracy and lessen time required for software fault root cause prediction as compared to conventional works. 

A Cooperative Spectrum Sensing Scheme using Particle Swarm Optimization and Cultural Algorithm

Anilkumar Dulichand Vishwakarma; Dr. Girish Ashok Kulkarni

European Journal of Molecular & Clinical Medicine, 2020, Volume 7, Issue 4, Pages 921-932

Cognitive radio is a software-based technology that provides dynamical access to unused or underused spectrum bands and enables spectrum sharing without causing any disadvantage among users. The performance of the cognitive radio in wireless communication networks depends on the accurate and fast detection of spectrum gaps.
The idea of Cognitive Radio is to share the spectrum between a so-called primary user and a so-called secondary user. The main objective of this spectrum management is to obtain a maximum rate of exploitation of the radio spectrum, for this cooperation between users is necessary. This paper provides a spectrum sensing approach using cooperation and competition to solve the spectrum allocation problem and thus ensure better management. The aim of spectrum detection is to detect spectrum gaps accurately and quickly. Therefore, the performance of cognitive radio networks largely depends on the spectrum sensing function. Particle Swarm Optimization (PSO) and Cultural Algorithm (CA) are proposed to increase spectrum detection performance in cognitive radio networks. The collaborative spectrum detection performance analysis of the proposed method was performed in Rayleigh fading channel in addition to the non-damped AWGN channel. As a result of simulation studies, it has been shown that a more effective perception can be made in cognitive radio networks by optimizing the threshold value expression from the historical data