Particle Swarm Optimization Feature Research Articles

Crop Mapping Using Random Forest and Particle Swarm Optimization based on Multi-Temporal Sentinel-2

Timely and accurate information on crop mapping and monitoring is necessary for agricultural resources management. Accordingly, the applicability of the proposed classification-feature selection ensemble procedure with different feature sets for crop mapping is investigated. Here, we produced various feature sets including spectral bands, spectral indices, variation of spectral index, texture, and combinations of features to map different types of crops. By using various feature sets and the random forest (RF) classifier, the crop maps were created. In aiming to determine the most relevant and distinctive features, the particle swarm optimization (PSO) and RF-variable importance measure feature selection methods were examined. The classification-feature selection ensemble procedure was adapted to combine the outputs of different feature sets from the better feature selection method using majority votes. Multi-temporal Sentinel-2 data has been used in Ghale-Nou county of Tehran, Iran. The performance of RF was efficient in crop mapping especially by spectral bands and texture in combination with other feature sets. Our results showed that the PSO-based feature selection leads to a more accurate classification than the RF-variable importance measure, in almost all feature sets for all crop types. The RF classifier-PSO ensemble procedure for crop mapping outperformed the RF classifier in each feature set with regard to the class-wise and overall accuracies (OA) (of about 2.7–7.4% increases in OA and 0.48–3.68% (silage maize), 0–1.61% (rice), 2.82–15.43% (alfalfa), and 10.96–41.13% (vegetables) improvement in F-scores for all feature sets). The proposed method could mainly be useful to differentiate between heterogeneous crop fields (e.g., vegetables in this study) due to their more obtained omission/commission errors reduction.

Determination of Customer Satisfaction of Tax Service Office Services Using C4.5 and PSO

Tax Service Office is a work unit of the Directorate General of Taxation that carries out services in the field of taxation to the public, both registered and unregistered taxpayers, within the working area of the Directorate General of Taxes. The number of Primary Tax Service Offices in Indonesia, one of which is the Primary Tax Service Office in Bekasi, has various ways to increase the satisfaction of taxpayers for the services provided. This study aims to determine the accuracy of taxpayers' satisfaction using data mining techniques using the Decision Tree C4.5 Algorithm with Particle Swarm Optimization (PSO) feature selection, validation uses cross validation techniques while accuracy is measured by the confussion matrix, which is to determine the level of service satisfaction conducted by distributing questionnaires to taxpayers in the Primary Tax Service Office in Bekasi as many as 500 questionnaires. The results show the accuracy value of Taxpayers' service satisfaction at the Pratama Tax Service Office using the Decision Tree C4.5 Algorithm with a feature selection of Particle Swarm Optimization (PSO) of 98,85%, Precission of 98,85% and Recall of 100%.

A Novel Hybrid PSO Assisted Optimization for Classification of Intellectual Disability Using Speech Signal

Speech signals convey speaker’s neurodevelopmental state along with phonological information. Recognize a speech disorder by analyzing the speech is essential for human–machine interaction. To develop a subject independent speech recognition system for neurodevelopmental disorders, by identifying voice features from MATLAB toolbox, spectral characteristics and feature selection algorithms are proposed in this paper. Feature selection is applied to overcome the challenges of dimensionality in various applications. This work presents a novel particle swarm optimization (PSO) based algorithm for feature selection. The experiments were conducted using a speech database of the children with intellectual disability with age-matched typically developed and validate the reliability using 10-fold cross-validation technique. The database consists of 141 speech features extracted from linear predictive coding (LPC) based cepstral parameters and Mel-frequency cepstral coefficients (MFCC). Three classification models were applied and obtained the recognition accuracies 90.30% with ANN, 98.00% with SVM and 91.00% with random forest with PSO feature selection algorithm. The results strongly prove the usefulness of the proposed multivariate feature selection algorithm when compared with filter approach.

COMPARATIVE ANALYSIS OF 5 ALGORITHM BASED PARTICLE SWARM OPTIMIZATION (PSO) FOR PREDICTION OF GRADUATE TIME GRADUATION

<strong>Graduation information is very important for Higher Education involved in education. The data received by students each year is an important part as a source of information for making decisions on the Higher Education side in admitting new students. The results show the PSO-based K-NN Algorithm at k-optimum = 19 has the best performance of the 5 existing algorithms, with an Accuracy value = 74.08% and an Under Curve Area (AUC) value = 0.788. Attributes of Gender, Semester Achievment Index 1, 2, 4, 6 and 7 as well as Employment Status make a real contribution to the right graduation of students. The addition of the Particle Swarm Optimization (PSO) feature always increases the accuracy value, while the highest increase in accuracy value in the Decision Tree (C4.5) Algorithm is 5.21%, the lowest in the Vector Support Engine Algorithm of 1.79%. The K-Nearest Neighbor (k-NN) algorithm corresponds to the third order, it remains the algorithm that has the best value, the highest accuracy value, this is due to the questionable value before discussing the PSO features. For the evaluation phase, the results of the accuracy are much better if using the overall data training (Angaktan 2007-2011), both with the 2011 Force test data or the 2010-2011 Force. Timely graduation of students begins in the class of 2011, with a value of 98.18% (100% resolution), meaning students start graduating on time</strong>

Analyzing cerebral infarction using support vector machine with artificial bee colony and particle swarm optimization feature selection

Early diagnosis of cerebral infarction is essential since many patients cannot be cured where the diagnosis is made at an advanced stage. In case an infarct occurs, the tissue in the brain die and stop the circulation of blood, which carries oxygen and nutrients to the body. Therefore, this study uses a machine learning Support Vector Machine (SVM) for early detection of the disorder. To produce the best classification accuracy and fast computing time, feature selection is performed on cerebral infarction data, including Artificial Bee Colony (ABC) and Particle Swarm Optimization (PSO). After classification, infarction data with the best features are classified using SVM. The classification results of ABC-SVM and PSO-SVM methods are compared with the accuracy of 90.36% for ABC-SVM and 86.74% for PSO-SVM. Therefore, the best approach used in classification is the SVM method with ABC feature selection.

Fuzzy Particle Swarm Optimization Feature Selection and Aggrandized Classifier for Uncovering Frauds in Credit Card Deals

In today's economy, credit card plays a very important role. The rise of credit card customers improved, credit card scam cases were also on the rise. Numerous procedures are anticipated to challenge the evolution of the frauds in credit cards. In this research work, proposed an innovative fraud detection method which utilizes the similar cardholder’s behavioral patterns to construct a current cardholder’s interactive profile in order to stay away from the credit card scams. However, the selection of optimal features from the samples and the decision cost for accuracy becomes main important problem. To illuminate these issues this proposed research work presents an innovative fraud detection technique that makes out of four phases: 1. To augment a cardholder’s behavioral styles, first we divide all cardholders into distinctive groups making use of the cardholder’s historical transaction data such that the members of each group have the similar transaction behavior by K-means. 2. Introduces a new Fuzzy Particle Swarm Optimization (FPSO) feature selection for the amplification of fraud detection in credit cards. 3. By means of a prolonged wrapper method, an ensemble classification are performed by Aggrandized Kernel based Support Vector Machine (AKSVM).4.Refreshing the cardholder’s social profile with an input system. This Proposed work adopts the external quality metrics as Accuracy, Recall, Concept drift rate and Fraud feature rate. The UCI dataset is used and is done in MATLAB framework. The analytical measures were used to estimate the routine of the mentioned fraud detection technique. The simulation results show that this proposed innovative fraud detection method provides better accuracy results than other fraud detection techniques. The low concept drift rate results the gain of the innovative method to classify the transactions accurately.

Detecting botnet by using particle swarm optimization algorithm based on voting system

Botnets have recently been identified as serious Internet threats that are continually developing and expanding. Identifying botnets in the domain of network security is regarded as a new challenge and topic for research. There are several methods for detecting botnets in networks, and prior research has encountered problems, including a high error and inaccuracy in detection. In this paper, the botnet detection method by using a hybrid of particle swarm optimization (PSO) algorithm with a voting system (BD-PSO-V) was used to improve the challenges of previous studies. The PSO algorithm was employed to select outstanding and effective features in the detection of botnets. The voting system, including a deep neural network algorithm, support vector machine (SVM), and decision tree C4.5, were utilized to identify botnets and classify samples. The decision-making strategy of the voting system was based on maximum votes, and the most important innovation of this research was to combine the PSO feature selection algorithm with a voting system using deep learning to identify botnets. Two datasets, ISOT and Bot-IoT, were employed to further verify the BD-PSO-V system performance. BD-PSO-V simulation improved the accuracy by an average of ∼0.42% and 0.17% in the ISOT dataset and the Bot-IoT dataset, respectively, compared to the other methods investigated. In addition, the effect of six well-known adversarial attacks on both datasets was evaluated. Despite a slight drop in accuracy rate, BD-PSO-V results had a promising performance against a variety of attacks.

A Hybrid Metaheuristic Approach for the Solution of Renewables-Incorporated Economic Dispatch Problems

This paper presents a hybrid metaheuristic optimization algorithm developed to solve the Economic Dispatch Problem (EDP) encountered in different combinations of power plants. The algorithm is developed by assimilating the prominent features of Particle Swarm Optimization (PSO) and Bat Algorithm (BA) and improves cost reduction and convergence with lesser computational time. The developed algorithm is employed for the solution of EDP consisting of only Renewable Energy Sources (RESs) implemented at various locations in Pakistan. The all RES based EDP consists of scenarios composed of sub-scenarios having no constraints, with time-varying loads and multi-area economic dispatch (MAED). The algorithm is also tested for three different combinations of power plants, comprising of RES integrated with thermal power plants (TPPs), the small-scaled thermal power system with constraints, and a large-scaled power system with Valve Point Loading (VPL) effect. The comparative analysis of the results for the developed metaheuristic algorithm with various existing techniques shows a reasonable reduction in the cost, improved computational time, and fast convergence.

A two-stage strategy for security-constrained AC dynamic transmission expansion planning

This paper presents a new and promising strategy organized in two stages to solve the dynamic multiyear transmission expansion planning, TEP, problem. Specifically, the first stage is related to the reduction of the search space size and it is conducted by a novel constructive heuristic algorithm (CHA). The second one is responsible for the refinement of the optimal solution plan and it uses a novel evolutionary algorithm based on the best features of particle swarm optimization (PSO) and genetic algorithm (GA). The planning problem is modelled as a dynamic and multiyear approach to ensure that it keeps a holistic view over the entire planning horizon and it aims at minimizing the total system costs comprising the investment and operation costs. Additionally, the N-1 contingency criterion is also considered in the problem. The developed approach was tested using the IEEE 118-Bus test system and the obtained results demonstrate its advantages in terms of efficiency and required computational time. Furthermore, the results demonstrated that the novel strategy can enable the utilization of the AC optimal power flow (OPF) in a faster and reliable way when compared to the standard and widespread DC-OPF model.

Sentiment Analysis Analysis of E-Wallet Sentiments on Google Play Using the Naive Bayes Algorithm Based on Particle Swarm Optimization

Increasingly sophisticated technology brings various conveniences both in transportation, information, education to the convenience of transactions in shopping, such as the development of E-wallet can now be easily done using a smartphone. From a number of e-wallet products, researchers took a case study, which is OVO product, which is currently being discussed by many groups, especially in the capital of Jakarta today. Customers or clients who are not satisfied with the services or products offered by a company will usually write their complaints on social media or reviews on Google play. However, monitoring and organizing opinions from the public is also not easy. For this reason, we need a special method or technique that is able to categorize these reviews automatically, whether positive or negative. The algorithm used in this study is Naive Bayes Classifier (NB), with the optimization of the use of Particle Swarm Optimization Feature Selection (FS). The results of cross validation NB without FS are 82.30% for accuracy and 0.780 for AUC. Whereas for NB with FS is 83.60% for accuracy and 0.801 for AUC. Very significant improvement with the use of Feature Selection (FS) Particle Swarm Optimization.

Heart disease detection using hybrid of bacterial foraging and particle swarm optimization

Bacterial-foraging-optimization (BFO) has newly raised and one of the most useful nature inspired optimization algorithm for real parametric optimization. During the process of random walk, the BFO algorithm makes search in the random direction, which increases delay. To overcome the delay in reaching the global optimum and also to boost up the performance of BFO, we proposed an algorithm by mixing the features of BFO and particle swarm optimization (PSO) for detecting the abnormal cardiac beat. Computer simulations illustrate the usefulness of the developed approach compared to the basic versions of BFO and PSO. The main aim of the research is to develop new modifications of BFO and its combination with transform technique such as Wavelet Transform and machine learning method, support vector machines (SVMs) to test their performances in the detection of cardiac arrhythmia. Modification of BFO focuses for improving its convergence in terms of speed and accuracy. Provided results in this paper show that, for the detection of MI and BBB classes, the BFPSO algorithm with SVM gives 98.9% and 99.3% accuracy on MIT-BIH database by including NSR database also. Moreover, the results demonstrate the effectiveness of the proposed method to improve the detection of cardiac arrhythmia.

Unsupervised feature selection based on bio-inspired approaches

In recent years, the scientific community has witnessed an explosion in the use of pattern recognition algorithms. However, little attention has been paid to the tasks preceding the execution of these algorithms, the preprocessing activities. One of these tasks is dimensionality reduction, in which a subset of features that improves the performance of the mining algorithm is located and algorithm's runtime is reduced. Although there are many methods that address the problems in pattern recognition algorithms, effective solutions still need to be researched and explored. Hence, this paper aims to address three of the issues surrounding these algorithms. First, we propose adapting a promising meta-heuristic called biased random-key genetic algorithm, which considers a random initial population construction. We call this algorithm as unsupervised feature selection by biased random-key genetic algorithm I. Next, we propose an approach for building the initial population partly in a deterministic way. Thus, we applied this idea in two algorithms, named unsupervised feature selection by particle swarm optimization and unsupervised feature selection by biased random-key genetic algorithm II. Finally, we simulated different datasets to study the effects of relevant and irrelevant attributes, and of noisy and missing data on the performance of the algorithms. After the Wilcoxon rank-sum test, we can state that the proposed algorithms outperform all other methods in different datasets. It was also observed that the construction of the initial population in a partially deterministic way contributed to the better performance. It should be noted that some methods are more sensitive to noisy and missing data than others, as well as to relevant and irrelevant attributes. • Particle swarm optimization and biased random-key genetic algorithms are proposed. • It is investigated how the construction of the initial population, as well as the missing and noisy data, implies the performance of the algorithms. • Ten simulated data sets are used containing relevant and irrelevant attributes, as well as different percentages of missing and noisy data. • An exhaustive computational and statistical evaluation is carried out. • The proposed approaches outperformed other methods.

Accelerated Simulated Annealing and Mutation Operator Feature Selection method for Big Data

The optimal feature subset selection over very high dimensional data is a vital issue. Even though the optimal features are selected, the classification of those selected features becomes a key complicated task. In order to handle these problems, a novel, Accelerated Simulated Annealing and Mutation Operator (ASAMO) feature selection algorithm is suggested in this work. For solving the classification problem, the Fuzzy Minimal Consistent Class Subset Coverage (FMCCSC) problem is introduced. In FMCCSC, consistent subset is combined with the K-Nearest Neighbour (KNN) classifier known as FMCCSC-KNN classifier. The two data sets Dorothea and Madelon from UCI machine repository are experimented for optimal feature selection and classification. The experimental results substantiate the efficiency of proposed ASAMO with FMCCSC-KNN classifier compared to Particle Swarm Optimization (PSO) and Accelerated PSO feature selection algorithms.

Improved Identification of Various Conditions of Induction Motor Bearing Faults

Artificial intelligence evolved as a powerful tool in condition monitoring of the induction motor for early diagnosis of bearing faults. This paper attempted to identify the features to improve the accuracy of diagnosis using the benchmark database of the Case Western Reserve University (CWRU) and the Machinery Fault Prevention Technology (MFPT). In this paper, 18 time-domain features are extracted constituting six proposed time-dependent spectral features (TDSFs) and 12 statistical time-domain features. For the first time, a set of six TDSFs are extracted to diagnose the bearing faults. This involves extracting frequency-domain features using the relationship between Parseval’s theorem and the Fourier transform directly from the time domain. Particle swarm optimization (PSO) and wheel-based differential evolution (WBDE) feature selection algorithms are also implemented to identify four prominent features. It is found that three of the four selected features are TDSF features, and results of selected features revealed the attainment of 90.92% for 48 class identifications of CWRU database and 100% for 17 class identifications of the MFPT database, respectively.

The cross-project defect prediction based on PSO and Feature Dependent Naive Bayes

The software defect prediction often assumes that the software under test has rich historical information, which is a harsh condition in reality. Cross-project defect prediction can predict projects with rare historical information by using historically informative projects. The differences in software architecture and code environment pose challenges to cross-project software defect prediction. Based on the characteristics of software defect prediction sets, this paper introduces a two-stage algorithm based on particle swarm optimization algorithm and Feature Dependent Naive Bayesian classifier.

PSO-FuzzyNN Techniques in Gender Classification Based on Bovine Bone Morphology Properties

This simulation project aims to solve forensic anthropology issues by using the computational method. The positive identification on gender is such a potential field to be explored. Basically, gender identification in forensic anthropology by comparative skeletal anatomy by atlas and crucially affect the identification accuracy. The simulation identification method was studied in order to determine the best model, which reduce the total costs of the post-mortem as an objective. The computational method on simulation run improves the identification accuracy as proven by many studies. Fuzzy K-nearest neighbours classifier (FuzzyNN) is such a computational intelligence method and always shows the best performance in many fields including forensic anthropology. Thus, this intelligent identification method was implemented within the determining for best accuracy. The result of this proposed model was compared with raw data collection and standard collections datasets; Goldman Osteometric dataset and Ryan and Shaw Dataset (RSD) as a benchmark for the identification policy. To improve the accuracy of FuzzyNN classifier, Particle Swarm Optimization (PSO) feature selection was used as the basis for choosing the best features to be used by the selected FuzzyNN classification model. The model is called PSO-FuzzyNN and has been developed by MATLAB and WEKA tools platform. Comparisons of the performance measurement namely the percentage of the classification accuracy of the model were performed. The result show potential the proposed PSO-FuzzyNN method demonstrates the capability to the obtained highest accuracy of identification.

Designing an Optimized Water Quality Monitoring Network with Reserved Monitoring Locations

The optimized design of water quality monitoring networks can not only minimize the pollution detection time and maximize the detection probability for river systems but also reduce redundant monitoring locations. In addition, it can save investments and costs for building and operating monitoring systems as well as satisfy management requirements. This paper aims to use the beneficial features of multi-objective discrete particle swarm optimization (MODPSO) to optimize the design of water quality monitoring networks. Four optimization objectives: minimum pollution detection time, maximum pollution detection probability, maximum centrality of monitoring locations and reservation of particular monitoring locations, are proposed. To guide the convergence process and keep reserved monitoring locations in the Pareto frontier, we use a binary matrix to denote reserved monitoring locations and develop a new particle initialization procedure as well as discrete functions for updating particle’s velocity and position. The storm water management model (SWMM) is used to model a hypothetical river network which was studied in the literature for comparative analysis of our work. We define three pollution detection thresholds and simulate pollution events respectively to obtain all the pollution detection time for all the potential monitoring locations when a pollution event occurs randomly at any potential monitoring locations. Compared to the results of an enumeration search method, we confirm that our algorithm could obtain the Pareto frontier of optimized monitoring network design, and the reserved monitoring locations are included to satisfy the management requirements. This paper makes fundamental advancements of MODPSO and enables it to optimize the design of water quality monitoring networks with reserved monitoring locations.

Performance Analysis of Particle Swarm Optimization for Feature Selection

One of the key task in data mining is the selection of relevant features from datasets with high dimensionality. This is expected to reduce the time and space complexity, and consequently improve the performance of data mining algorithms for tasks such as classification. This study presents an empirical study of the effect of particle swarm optimization as a feature selection technique on the performance of classification algorithms. Two dataset from different domains were used: SMS spam detection and sentiment analysis datasets. Particle swarm optimization is applied on the datasets for feature selection. Both the reduced and raw dataset are separately classified using C4.5 decision tree, k-nearest neighbour and support vector machine. The result of the analysis showed that the improvement of classifier performance is case-dependent; some significant improvements are noticed in the sentiment analysis datasets and not in the SMS spam dataset. Although some marginal effect are observed on performance, it implies that with particle swarm optimization features selection the space complexity is reduced while maintaining the accuracy of the classifiers. Keywords—classification, feature selection, machine learning, particle swarm optimization, text mining

Asymmetric Bouc–Wen hysteresis modeling and inverse compensation for piezoelectric actuator via a genetic algorithm–based particle swarm optimization identification algorithm

The hysteresis characteristics, which commonly existed in smart materials–based actuators, play a significant role in precision control technology. In this article, a modified Bouc–Wen model which can describe the asymmetric hysteresis characteristics of piezoelectric ceramic actuators is investigated. The corresponding parameters of the modified Bouc–Wen hysteresis model are identified through a genetic algorithm–based particle swarm optimization algorithm. Compared with independent particle swarm optimization method which is easily trapped in the local extremum, the proposed genetic algorithm–based particle swarm optimization features the strong searching ability both in early global search period and the later local search period. The experimental results show that the asymmetric Bouc–Wen model identified via genetic algorithm–based particle swarm optimization algorithm are more accurate than that identified through independent particle swarm optimization or genetic algorithm approach, and the maximum displacement error and the maximum relative error between the genetic algorithm–based particle swarm optimization model and the experimental value are 0.20 µm and 14.28%, respectively, which are much smaller than that of particle swarm optimization method with 0.67 µm and 47.85% and genetic algorithm method with 0.35 µm and 25%. In order to further verify the accuracy of the identified model, the hysteresis compensation of piezoelectric ceramic actuator was realized using the feedforward controller based on the inverse Bouc–Wen model.

Recognition of Shooter’s Emotions Under Stress Based on Affective Computing

This paper attempts to accurately recognize the emotional states of shooters and prepare practical adjustment strategies for these athletes. First, an experimental paradigm was designed for emotion induction, which effectively arouses the calmness, sadness, fear, and happiness, four common emotions, of the subjects with emotion-inducing video clips. The electrodermal activities (EDAs) collected under the four emotions were employed to analyze the emotions of the subjects, and six eigenvectors were identified by particle swarm optimization (PSO) to effectively recognize stress emotions. Then, the k-nearest neighbor's algorithm (kNNs) was adopted to calculate the emotions under stress and recognize the stress degree. To improve recognition accuracy, the captured emotions were subjected to baseline removal and PSO feature optimization. Furthermore, an emotional model for the athletes was set up based on the emotional probability space of the Markov chain (MC). This model can accurately simulate the emotional features and emotional states of the subjects. The research results provide a good reference for the objective evaluation of athletes' emotions and the maintenance of stable and proper emotional states of shooters in daily training and competition.