Fast Laplacian twin support vector machine with active learning for pattern classification

In this paper, we propose two Laplacian nonparallel hyperplane proximal classifiers (LapNPPCs) for semi-supervised and full-supervised classification problem respectively by adding manifold regularization terms. Due to the manifold regularization terms, our LapNPPCs are able to exploit the intrinsic structure of the patterns of the training set. Furthermore, our classifiers only need to solve two systems of linear equations rather than two quadratic programming (QP) problems as needed in Laplacian twin support vector machine (LapTSVM) (Z. Qi, Y. Tian and Y. Shi, Neural Netw.35 (2012) 46–53). Numerical experiments on toy and UCI benchmark datasets show that the accuracy of our LapNPPCs is comparable with other classifiers, such as the standard SVM, TWSVM and LapTSVM, etc. It is also the case that based on our LapNPPCs, some other TWSVM type classifiers with manifold regularization can be constructed by choosing different norms and loss functions to deal with semi-supervised binary and multi-class classification problems.

Laplacian least squares twin support vector machine for semi-supervised classification

Network security has become one of the well-known concerns in the last decades. Machine learning techniques are robust methods in detecting malicious activities and network threats. Most previous works learn offline supervised classifiers while they require large amounts of labeled examples and also should update models because the data change over time in real world applications. To alleviate these problems, we propose a novel online version of laplacian twin support vector machine classifier, which can exploit the geometry information of the marginal distribution embedded in unlabeled data to construct a more accurate and faster semi-supervised classifier. The results of experiments on large network datasets show that Online Lap-TSVM combined by two nonparallel hyper planes improves the accuracy with the comparable computing time and storage to Lap-TSVM.

Over the past decades, 1-norm techniques based on algorithms are widely used to suppress input features. Quite different from traditional 1-norm support vector machine (SVM), direct 1-norm optimization based on the primal problem of nonparallel plane classifiers like generalized proximal support vector machine, twin support vector machine (TWSVM) and least squares twin support vector machine (LSTSVM) are not capable of generating very sparse solutions that are vital for classification and can make them easier to store and faster to compute. To address the issue, in this paper, we develop a feature selection method for LSTSVM, called a feature selection method for nonparallel plane support vector machine classification (FLSTSVM), which is specially designed for strong feature suppression. We incorporate a Tikhonov regularization term to the objective of LSTSVM, and then minimize its 1-norm measure. Solution of FLSTSVM can follow directly from solving two smaller quadratic programming problems (QPPs) arising from two primal QPPs as opposed to two dual ones in TWSVM. FLSTSVM is capable of generating very sparse solutions. This means that FLSTSVM can reduce input features, for the linear case. When a nonlinear classifier is used, few kernel functions determine the classifier. In addition to having strong feature suppression, the edge of our method still lies in its faster computing time compared to that of TWSVM, Newton Method for Linear Programming SVM (NLPSVM) and LPNewton. Lastly, this algorithm is compared on public data sets, as well as an Exclusive Or (XOR) example.

Gene expression profiles have been used to predict cancer recurrence or other clinical outcomes of cancer patients. However, clinical information of cancer patients is often incomplete, which yields many unlabeled samples that cannot be used in supervised learning. In this is paper, we develop a novel semi-supervised leaning (SSL) method that uses both labeled and unlabeled patient samples to predict cancer recurrence. Our new SSL algorithm employs a sparse representation approach where a labeled sample is represented as a combination of a small number of properly chosen unlabeled samples. Experiments with a set of gene expression data from patients with colorectal cancer(CRC) demonstrate that our SSL algorithm can improve prediction accuracy compared to other two SSL methods including TSVM and T3VM, and the traditional support vector machine.

Detection of credit card fraud has lately been considered a critical task due to the highly imbalanced nature of financial transaction databases. On the other hand, the traditional classification algorithms have been poor at detecting fraudulent activities with an acceptable false-positive rate. Hence, this work contributes to a GMM based approach for fraud detection, which benefits from GMM probabilistic-based classification power for better classification results. The database used in this study is publicly available and was obtained from the Kaggle Credit Card Fraud Detection (CCFD) database. The dataset has 284,807 transactions, and only 0.17% of the cases represent fraud. This paper plans to scale the features, train the GMM technique with different numbers of Gaussian components (i.e., 2, 4, 6, 8, and 10), and evaluate their performance with several evaluation metrics. Compared with other traditional classifiers (logistic regression (92.4%), K-nearest neighbors (93.68%), decision tree (88.16%), and support vector machine (94.21%)), the proposed GMM algorithm obtains the highest accuracy of 94.53%. The proposed method, despite its high accuracy, has limitations under high-dimensional feature dependencies and optimal component selection. From the results obtained over the experimentation process, the GMM proves to be a probable, yet flexible and subservient framework for a complex modelling of probabilities for the detection of fraud.

Customer satisfaction and their positive sentiments are some of the various goals for successful companies. However, analyzing customer reviews to predict accurate sentiments have been proven to be challenging and time-consuming due to high volumes of collected data from various sources. Several researchers approach this with algorithms, methods, and models. These include machine learning and deep learning (DL) methods, unigram and skip-gram based algorithms, as well as the Artificial Neural Network (ANN) and bag-of-word (BOW) regression model. Studies and research have revealed incoherence in polarity, model overfitting and performance issues, as well as high cost in data processing. This experiment was conducted to solve these revealing issues, by building a high performance yet cost-effective model for predicting accurate sentiments from large datasets containing customer reviews. This model uses the fastText library from Facebook’s AI research (FAIR) Lab, as well as the traditional Linear Support Vector Machine (LSVM) to classify text and word embedding. Comparisons of this model were also done with the author’s a custom multi-layer Sentiment Analysis (SA) Bi-directional Long Short-Term Memory (SA-BLSTM) model. The proposed fastText model, based on results, obtains a higher accuracy of 90.71% as well as 20% in performance compared to LSVM and SA-BLSTM models.

A sparse extreme learning machine framework by continuous optimization algorithms and its application in pattern recognition

This study presents a cross-comparison of traditional linear stochastic estimation (LSE) and support vector machine (SVM) algorithms. The assessment of the capabilities of both estimation techniques is based on reconstructing the unsteady behavior of a laminar separation bubble (LSB) on a NACA 0018 airfoil at a Reynolds number of 100,000. The algorithms are trained based on time-resolved velocity field measurements performed simultaneously with sparse unsteady surface pressure measurements. The flow reconstructions performed based on surface pressure measurements are evaluated based on an independent set of flow field measurements. The results show a comparable performance across multi-point LSE and different SVM methods investigated, which are shown to capture the flow development of dominant coherent structures in the separated shear layer.

Customer satisfaction and their positive sentiments are some of the various goals for successful companies. However, analyzing customer reviews to predict accurate sentiments have been proven to be challenging and time-consuming due to high volumes of collected data from various sources. Several researchers approach this with algorithms, methods, and models. These include machine learning and deep learning (DL) methods, unigram and skip-gram based algorithms, as well as the Artificial Neural Network (ANN) and bag-of-word (BOW) regression model. Studies and research have revealed incoherence in polarity, model overfitting and performance issues, as well as high cost in data processing. This experiment was conducted to solve these revealing issues, by building a high performance yet cost-effective model for predicting accurate sentiments from large datasets containing customer reviews. This model uses the fastText library from Facebook’s AI research (FAIR) Lab, as well as the traditional Linear Support Vector Machine (LSVM) to classify text and word embedding. Comparisons of this model were also done with the author’s a custom multi-layer Sentiment Analysis (SA) Bi-directional Long Short-Term Memory (SA-BLSTM) model. The proposed fastText model, based on results, obtains a higher accuracy of 90.71% as well as 20% in performance compared to LSVM and SA-BLSTM models.

Drug classification and target identification are crucial yet challenging steps in drug discovery. Existing methods often suffer from inefficiencies, overfitting, and limited scalability. Traditional approaches like support vector machines and XGBoost struggle to handle large, complex pharmaceutical datasets effectively. Deep learning models, while powerful, face challenges with interpretability, computational complexity, and generalization to unseen data. This study addresses these limitations by introducing a novel framework: optSAE+HSAPSO. This framework integrates a stacked autoencoder (SAE) for robust feature extraction with a hierarchically self-adaptive particle swarm optimization (HSAPSO) algorithm for adaptive parameter optimization. This combination delivers superior performance across various classification metrics. Experimental evaluations on datasets from DrugBank and Swiss-Prot demonstrate that optSAE+HSAPSO achieves a high accuracy of 95.52%. Notably, it exhibits significantly reduced computational complexity (0.010 seconds per sample) and exceptional stability (±0.003). Compared to state-of-the-art methods, the framework offers higher accuracy, faster convergence, and greater resilience to variability. Furthermore, ROC and convergence analyses confirm its robustness and generalization capability, maintaining consistent performance across both validation and unseen datasets. By leveraging advanced optimization techniques, the framework efficiently handles large feature sets and diverse pharmaceutical data, making it a scalable and adaptable solution for real-world drug discovery applications. However, the method's performance is dependent on the quality of the training data, and fine-tuning may be necessary for high-dimensional datasets. Despite these limitations, the optSAE+HSAPSO framework demonstrates transformative potential, significantly reducing computational overhead while improving classification accuracy and reliability. This work advances the field of pharmaceutical informatics by presenting a reliable and efficient framework for drug classification and target identification. These findings open promising avenues for future research, including extending the framework to other domains such as disease diagnostics or genetic data classification, ultimately accelerating the drug development process.

Traditional hinge loss function-based large-scale support vector machine (SVM) algorithms tend to perform poorly in the presence of noise, especially when the model is trained incrementally. In this paper, we propose an efficient stochastic quasi-Newton method-based twin parametric SVM using the pinball loss function (termed as SQN-PTWSVM), which is efficient and more robust to the presence of noise when compared to conventional hinge loss SVM for large-scale data scenarios. To establish the theoretical convergence of the method, a modified version of SQN-PTWSVM, termed as SQN-SPTWSVM, has also been proposed. It overcomes the poor convergence issue faced by stochastic gradient twin SVM thus resulting in a faster and reliable model. In SQN-SPTWSVM, the hyperplanes obtained are stable enough to handle noise and resampling issues that occur frequently in stochastic learning scenarios, leading to better generalization ability of the classifier. The proposed method has been extended to nonlinear scenarios as well. Moreover, batch versions of the proposed algorithms have also been introduced which significantly reduce the training time and memory requirement of SQN-PTWSVM and SQN-SPTWSVM. The experimental results on several benchmark datasets and activity recognition applications have shown that the performance of our method is better than the existing classifiers in terms of speed and accuracy.

To solve the problem that traditional multi-label support vector machine (SVM) classification algorithm adopting nonlinear kernel has been severely restricted from being used on large-scale data sets, we propose fast multi-label low-rank-linearized SVM classification algorithm based on approximate extreme points (AEML-LLSVM). First, it adopts the approximate extreme points' method to obtain representative sets from the training data set. Then, the approximate extreme points' low-rank-linearized SVM (AELLSVM) is trained on the representative sets. The AELLSVM integrates the advantages of approximate extreme points' method and LLSVM. Experimental results on three large-scale multi-label data sets have proven that the training and the testing speed of AEML-LLSVM classification algorithm are greatly improved under the premise that its classification performance is similar to that of ML-LIBSVM classification algorithm and superior to that of other fast multi-label SVM classification algorithms.

In order to fully utilize the local geometric information of the given training set consisting of the normal data, locality correlation preserving (LCP) is introduced into the traditional one-class support vector machine (OCSVM). The proposed method, named as locality correlation preserving based one-class support vector machine (LCP-OCSVM), inherits the merits of LCP and OCSVM. It can keep locality correlation of the normal data and margin maximization between the normal data and the origin in the high-dimensional feature space. Experimental results on one synthetic data set and ten benchmark data sets demonstrate that the proposed method is superior to the traditional OCSVM and two related approaches.

An accurate computational approach for the prediction of pan evaporation over daily time horizons is a useful decisive tool in sustainable agriculture and hydrological applications, particularly in designing the rural water resource systems, water use allocations, utilization and demand assessments, and the management of irrigation systems. In this study, a hybrid predictive model (Multilayer Perceptron-Firefly Algorithm (MLP-FFA)) based on the FFA optimizer that is embedded within the MLP technique is developed and evaluated for its suitability for the prediction of daily pan evaporation. To develop the hybrid MLP-FFA model, the pan evaporation data measured between 2012 and 2014 for two major meteorological stations (Talesh and Manjil) located at Northern Iran are employed to train and test the predictive model. The ability of the hybrid MLP-FFA model is compared with the traditional MLP and support vector machine (SVM) models. The results are evaluated using five performance criteria metrics: root mean square error (RMSE), mean absolute error (MAE), Nash-Sutcliffe efficiency (NS), and the Willmott’s Index (WI). Taylor diagrams are also used to examine the similarity between the observed and predicted pan evaporation data in the test period. Results show that an optimal MLP-FFA model outperforms the MLP and SVM model for both tested stations. For Talesh, a value of WI = 0.926, NS = 0.791, and RMSE = 1.007 mm day−1 is obtained using MLP-FFA model, compared with 0.912, 0.713, and 1.181 mm day−1 (MLP) and 0.916, 0.726, and 1.153 mm day−1 (SVM), whereas for Manjil, a value of WI = 0.976, NS = 0.922, and 1.406 mm day−1 is attained that contrasts 0.972, 0.901, and 1.583 mm day−1 (MLP) and 0.971, 0.893, and 1.646 mm day−1 (SVM). The results demonstrate the importance of the Firefly Algorithm applied to improve the performance of the MLP-FFA model, as verified through its better predictive performance compared to the MLP and SVM model.