Subset Selection Technique Research Articles

An optimal feature subset selection technique to improve accounting information security for intrusion detection systems

An optimal feature subset selection technique to improve accounting information security for intrusion detection systems

Enhancing Adaboost performance in the presence of class-label noise: A comparative study on EEG-based classification of schizophrenic patients and benchmark datasets

The performance of Adaboost is highly sensitive to noisy and outlier samples. This is therefore the weights of these samples are exponentially increased in successive rounds. In this paper, three novel schemes are proposed to hunt the corrupted samples and eliminate them through the training process. The methods are: I) a hybrid method based on K-means clustering and K-nearest neighbor, II) a two-layer Adaboost, and III) soft margin support vector machines. All of these solutions are compared to the standard Adaboost on thirteen Gunnar Raetsch’s datasets under three levels of class-label noise. To test the proposed method on a real application, electroencephalography (EEG) signals of 20 schizophrenic patients and 20 age-matched control subjects, are recorded via 20 channels in the idle state. Several features including autoregressive coefficients, band power and fractal dimension are extracted from EEG signals of all participants. Sequential feature subset selection technique is adopted to select the discriminative EEG features. Experimental results imply that exploiting the proposed hunting techniques enhance the Adaboost performance as well as alleviating its robustness against unconfident and noisy samples over Raetsch benchmark and EEG features of the two groups.

Hybrid ML-MDKL feature subset selection and classification technique accompanied with rat swarm optimizer to classify the multidimensional breast cancer mammogram image

Breast cancer is a pathological condition characterized by the abnormal proliferation of cells inside the breast tissue. The biggest challenge at the moment is determining the most optimal subset from large datasets. Numerous methodologies have been presented for the purpose of selecting the most suitable subset from datasets with high dimensions. However, the outcomes have proven to be insufficient in effectively addressing extensive quantities of multidimensional datasets. This publication presents a proposal for an efficient approach of selecting numerous feature subsets and performing classification on Multidimensional Datasets (MDD). In this study, the Moment Invariant Wavelet Feature Extraction (MI-WFE) approach is employed for the purpose of feature extraction. The New Adaptive Hybrid Levy Flight based Cuckoo Search Optimization Algorithm is employed to identify the most pertinent subset of features. This study presents a novel approach, referred to as H-RS-LVCSO, which aims to enhance the local search capability and optimization speed of the algorithm. This is achieved through the hybridization of the Rat Swarm Optimizer with the levy flight based cuckoo search optimization technique. In this study, it is suggested a hybridized approach for classifying breast cancer, utilizing a Multilayer Multiple Deep Kernel Learning (ML-MDKL) Classifier in conjunction with the standard Support Vector Machine (SVM) Classifier. The simulation of the suggested method is conducted using the MATLAB software. The approach under consideration is evaluated in comparison to two pre-existing methods. The suggested method demonstrates an accuracy improvement of 10.58% compared to existing methods. Furthermore, it outperforms these methods by 17.5% in terms of accuracy. The proposed method exhibits a precision improvement of 6.52% and 14.23% compared to the existing methods.

Multi-label learning by extended multi-tier stacked ensemble method with label correlated feature subset augmentation

<p>Classification is one of the basic and most important operations that can be used in data science and machine learning applications. Multi-label classification is an extension of the multi-class problem where a set of class labels are associated with a particular instance at a time. In a multiclass problem, a single class label is associated with an instance at a time. However, there are many different stacked ensemble methods that have been proposed and because of the complexity associated with the multi-label problems, there is still a lot of scope for improving the prediction accuracy. In this paper, we are proposing the novel extended multi-tier stacked ensemble (EMSTE) method with label correlation<br />by feature subset selection technique and then augmenting those feature subsets while constructing the intermediate dataset for improving the prediction accuracy in the generalization phase of the stacking. The performance effect of the proposed method has been compared with existing methods and showed that our proposed method outperforms the other methods.</p>

Metaheuristics with Deep Learning Model for Cybersecurity and Android Malware Detection and Classification

Since the development of information systems during the last decade, cybersecurity has become a critical concern for many groups, organizations, and institutions. Malware applications are among the commonly used tools and tactics for perpetrating a cyberattack on Android devices, and it is becoming a challenging task to develop novel ways of identifying them. There are various malware detection models available to strengthen the Android operating system against such attacks. These malware detectors categorize the target applications based on the patterns that exist in the features present in the Android applications. As the analytics data continue to grow, they negatively affect the Android defense mechanisms. Since large numbers of unwanted features create a performance bottleneck for the detection mechanism, feature selection techniques are found to be beneficial. This work presents a Rock Hyrax Swarm Optimization with deep learning-based Android malware detection (RHSODL-AMD) model. The technique presented includes finding the Application Programming Interfaces (API) calls and the most significant permissions, which results in effective discrimination between the good ware and malware applications. Therefore, an RHSO based feature subset selection (RHSO-FS) technique is derived to improve the classification results. In addition, the Adamax optimizer with attention recurrent autoencoder (ARAE) model is employed for Android malware detection. The experimental validation of the RHSODL-AMD technique on the Andro-AutoPsy dataset exhibits its promising performance, with a maximum accuracy of 99.05%.

Gene selection of microarray data using Heatmap Analysis and Graph Neural Network

It is not feasible to investigate the whole genes at a microscopic level for disease classification in Genomics. It might take substantial time to execute any meaningful analysis and the computational resources will be misused as not all the genes are responsible for the disease linked to a cell. Currently, it is quite challenging to select the most significant genes from high-dimensional microarray data for disease classification. In search of a better process, a novel gene subset selection technique has been developed based on Heatmap Analysis and Graph Neural Network (HAGNN). In the proposed method, a heatmap analysis has been performed for the different classes of microarray data to obtain the Region of Interest (ROIs). These ROIs are extracted from the original dataset and undergo a node reduction technique followed by an edge reduction technique in Graph Neural Network (GNN). This paper is concluded with an optimal subset of the most significant genes that cause cancer. The popular base classifiers have been used to evidence the importance of the selected genes as compared to the original data with the help of several metrics. The obtained results clearly show that the proposed methodology outperformed the other existing methods and make a greater impact on the advancement of the GNN-based gene selection method.

Hybrid Feature Selection based on BTLBO and RNCA to Diagnose the Breast Cancer

Feature selection is a feasible solution to improve the speed and performance of machine learning models. Optimization algorithms are doing a significant job in searching for optimal variables from feature space. Recent feature selection methods are purely depending on various meta heuristic algorithms for searching a good combination of features without considering the importance of individual features, which makes classification models to suffer from local optima or overfitting problems. In this paper, a novel hybrid feature subset selection technique is introduced based on Regularized Neighborhood Component Analysis (RNCA) and Binary Teaching Learning Based Optimization (BTLBO) algorithms to overcome the above problems. RNCA algorithm assigns weights to the attributes based on their contribution in building the learning models for classification. BTLBO algorithm computes the fitness of individuals with respect to the weights of features and selects the best ones. The results of similar feature selection methods are matched with the proposed hybrid model and proved better performance in terms of classification accuracy, recall and AUC measures over breast cancer datasets.

Deep Learning Approaches for Cyberbullying Detection and Classification on Social Media.

As a result of the ease with which the internet and cell phones can be accessed, online social networks (OSN) and social media have seen a significant increase in popularity in recent years. Security and privacy, on the other hand, are the key concerns in online social networks and other social media platforms. On the other hand, cyberbullying (CB) is a serious problem that needs to be addressed on social media platforms. Known as cyberbullying (CB), it is defined as a repetitive, purposeful, and aggressive reaction performed by individuals through the use of information and communication technology (ICT) platforms such as social media platforms, the internet, and cell phones. It is made up of hate messages that are sent by e-mail, chat rooms, and social media platforms, which are accessed through computers and mobile phones. The detection and categorization of CB using deep learning (DL) models in social networks are, therefore, crucial in order to combat this trend. Feature subset selection with deep learning-based CB detection and categorization (FSSDL-CBDC) is a novel approach for social networks that combines deep learning with feature subset selection. The suggested FSSDL-CBDC technique consists of a number of phases, including preprocessing, feature selection, and classification, among others. Additionally, a binary coyote optimization (BCO)-based feature subset selection (BCO-FSS) technique is employed to select a subset of features that will increase classification performance by using the BCO algorithm. Additionally, the salp swarm algorithm (SSA) is used in conjunction with a deep belief network (DBN), which is known to as the SSA-DBN model, to detect and characterize cyberbullying in social media networks and other online environments. The development of the BCO-FSS and SSA-DBN models for the detection and classification of cyberbullying highlights the originality of the research. A large number of simulations were carried out to illustrate the superior classification performance of the proposed FSSDL-CBDC technique. The SSA-DBN model has exhibited superior accuracy to the other algorithms, with a 99.983 % accuracy rate. Overall, the experimental results revealed that the FSSDL-CBDC technique beats the other strategies in a number of different aspects.

Adaptive Rank-Based Selection of Geological Realizations for Optimum Field Development Planning

SummaryUncertainties are present in many decision-making processes. In field development planning, these uncertainties, typically represented by a set of geological realizations, need to be propagated in response to any proposed alternative (solution). Incorporation of the full set of realizations results in the oil and gas field development optimization problem—where either an algorithm iteratively tries to find the best solution from all the possible alternatives or the best solution must be selected from a set of predefined engineering judgment-driven development scenarios (i.e., set of either well control or well placement settings)—becoming computationally demanding. As such, realization subset selection techniques are required to reduce the computational overhead. We first introduce a reformulation of the subset selection problem to one that aims at ensuring consistent ranking of alternatives between those obtained by the full set and the selected subset. We argue that this should be the ultimate goal of any subset selection technique in such problems. In addition, we also propose a technique which selects a subset that minimizes the difference between the rankings obtained by the full set and subset, for a small batch of alternatives. The key idea, which we investigate thoroughly, is that there is a positive association between the goodness (in terms of ranking alternatives) of the subset selected using a small batch of alternatives and its fidelity in ranking other alternatives. Unlike previous methods, this technique does not depend on selecting subjective (static) properties to perform the subset selection nor does it rely only on flow-response vectors of a base-case scenario. In this work, the proposed technique is assessed using well placement and well control development alternatives to determine the applicability within field development planning. Additionally, the proposed subset selection technique is implemented in an adaptive scheme to solve a well placement optimization problem. The results are promising as the proposed technique consistently selects subsets that are able to rank development alternatives in a similar manner to the full set regardless of the type of development strategy (well control settings or well placement). Furthermore, the implementation of the proposed technique in an adaptive scheme is able to reduce the computational costs, on average, by a factor close to 9 without compromising the solution found for well placement optimization.

Multi‐view rank‐based random forest: A new algorithm for prediction in eSports

AbstractThe main problem associated with the random forest (RF) algorithm is its application of random feature subset selection technique over a single vector. In this technique, the irrelevant or redundant features in a single‐view data have equal chances with the important features to be selected for training a classifier, leading to misclassification. To overcome this problem, this article proposes a novel algorithm, called multi‐view rank‐based random forest (MVRRF). The proposed algorithm builds a set of decision trees from multi‐view data by using a rank‐based feature selection strategy. The main advantages of our method are that (i) it extends the RF algorithm for multi‐view learning, and (ii) it reduces the chances of irrelevant and redundant feature selection, and thus it usually improves the accuracy, generalizability and robustness of the classification models. The aim of our study is to predict the match result of the game League of Legends in electronic sports (eSports). Thus, the eSports teams can define trustworthy strategies through important features. The proposed method can be successfully used in other fields as well as eSports. The experiments that were conducted on a publicly available eSports dataset show that the proposed MVRRF algorithm (93.32%) outperforms the standard RF algorithm (86.38%) on multi‐view data in terms of accuracy. Furthermore, the experimental results also show that our method achieved higher performance than the methods tested in the state of the art studies on the same dataset.

Predicting zebrafish (Danio rerio) embryo developmental toxicity through a non-conformational QSAR approach

For many years, the frequent use of synthetic chemicals in the manufacture of veterinary drugs and plague control products has raised negative effects on human health and other non-target organisms, promoting the need to employ a practical and suitable methodology for early risk identification of several thousand commercial compounds. The zebrafish (Danio rerio) embryo has been emerged as one sustainable animal model for measuring developmental toxicity, an endpoint that is included in the regulatory procedures to approve chemicals, avoiding conventional and costly toxicity assays based on animal testing. In this context, the Quantitative Structure-Activity Relationships (QSAR) theory is applied to develop a predictive model based on a well-defined zebrafish embryo developmental toxicity database reported by the ToxCast™ Phase I chemical library of the Environmental Protection Agency (U.S. EPA). By means of four freely available softwares, a set with 28,038 non-conformational descriptors that encode the largest amount of permanent structural features are readily calculated. The Replacement Method (RM) variable subset selection technique provided the best regression models. Thereby, a linear QSAR model with proper statistical quality (Rtrain2 = 0.64, RMSEtrain = 0.49) is established in agreement with the Organization for Economic Co-operation and Development principles, accomplishing each internal (loo, l15 % o, VIF and Y-randomization) and external (Rtest2,Rm2, QF12, QF22, QF32 and CCC) validation criterion. The present QSAR approach provides a useful computational tool to estimate zebrafish developmental toxicity of new, untasted or hypothetical compounds, and it can contribute to the general lack of QSAR models in the literature to predict this endpoint.

Alternative QSAR Study for Unsymmetrical Aromatic Disulfide Anti-SARS Inhibitors

This work applied the quantitative structure-activity relationships (QSAR) theory to predict the inhibitory activity exhibited by 40 unsymmetrical aromatic disulfide compounds against the SARS-CoV main protease. Different freely available molecular descriptor programs provided 67,116 independent non-conformational molecular descriptors. This great number of descriptors contained multidimensional representations of the chemical structure and was analyzed through multivariable linear regressions and the replacement method variable subset selection technique. The developed QSAR model achieved an acceptable statistical quality and provided a prospective guide that was considered useful for predicting the inhibitory activity of structurally-related aromatic disulfide compounds on the SARS-CoV main protease.

Modeling and Parameter Subset Selection for Fibrin Polymerization Kinetics with Applications to Wound Healing.

During the hemostatic phase of wound healing, vascular injury leads to endothelial cell damage, initiation of a coagulation cascade involving platelets, and formation of a fibrin-rich clot. As this cascade culminates, activation of the protease thrombin occurs and soluble fibrinogen is converted into an insoluble polymerized fibrin network. Fibrin polymerization is critical for bleeding cessation and subsequent stages of wound healing. We develop a cooperative enzyme kinetics model for in vitro fibrin matrix polymerization capturing dynamic interactions among fibrinogen, thrombin, fibrin, and intermediate complexes. A tailored parameter subset selection technique is also developed to evaluate parameter identifiability for a representative data curve for fibrin accumulation in a short-duration in vitro polymerization experiment. Our approach is based on systematic analysis of eigenvalues and eigenvectors of the classical information matrix for simulations of accumulating fibrin matrix via optimization based on a least squares objective function. Results demonstrate robustness of our approach in that a significant reduction in objective function cost is achieved relative to a more ad hoc curve-fitting procedure. Capabilities of this approach to integrate non-overlapping subsets of the data to enhance the evaluation of parameter identifiability are also demonstrated. Unidentifiable reaction rate parameters are screened to determine whether individual reactions can be eliminated from the overall system while preserving the low objective cost. These findings demonstrate the high degree of information within a single fibrin accumulation curve, and a tailored model and parameter subset selection approach for improving optimization and reducing model complexity in the context of polymerization experiments.

Metric and Accuracy Ranked Feature Inclusion: Hybrids of Filter and Wrapper Feature Selection Approaches

Feature selection has emerged as a craft, using which we boost the performance of our learning model. Feature or Attribute Selection is a data preprocessing technique, where only the most informative features are considered and given to the predictor. This reduces the computational overhead and improves the correctness of the classifier. Attribute Selection is commonly carried out by applying some filter or by using the performance of the learning model to gauge the quality of the attribute subset. Metric Ranked Feature Inclusion and Accuracy Ranked Feature Inclusion are the two novel hybrid feature selection methods we introduce in this paper. These algorithms follow a two-stage procedure, the first of which is feature ranking, followed by feature subset selection. They differ in the way they rank the features but follow the same subset selection technique. Multiple experiments have been conducted to assess our models. We compare our results with numerous works of the past and validate our models using 12 datasets. From the results, we infer that our algorithms perform better than many existent models.

Design and Development of an Efficient Network Intrusion Detection System Using Machine Learning Techniques

Today’s internets are made up of nearly half a million different networks. In any network connection, identifying the attacks by their types is a difficult task as different attacks may have various connections, and their number may vary from a few to hundreds of network connections. To solve this problem, a novel hybrid network IDS called NID‐Shield is proposed in the manuscript that classifies the dataset according to different attack types. Furthermore, the attack names found in attack types are classified individually helping considerably in predicting the vulnerability of individual attacks in various networks. The hybrid NID‐Shield NIDS applies the efficient feature subset selection technique called CAPPER and distinct machine learning methods. The UNSW‐NB15 and NSL‐KDD datasets are utilized for the evaluation of metrics. Machine learning algorithms are applied for training the reduced accurate and highly merit feature subsets obtained from CAPPER and then assessed by the cross‐validation method for the reduced attributes. Various performance metrics show that the hybrid NID‐Shield NIDS applied with the CAPPER approach achieves a good accuracy rate and low FPR on the UNSW‐NB15 and NSL‐KDD datasets and shows good performance results when analyzed with various approaches found in existing literature studies.

Histogram-Based Descriptor Subset Selection for Visual Recognition of Industrial Parts

This article deals with the 2D image-based recognition of industrial parts. Methods based on histograms are well known and widely used, but it is hard to find the best combination of histograms, most distinctive for instance, for each situation and without a high user expertise. We proposed a descriptor subset selection technique that automatically selects the most appropriate descriptor combination, and that outperforms approach involving single descriptors. We have considered both backward and forward mechanisms. Furthermore, to recognize the industrial parts a supervised classification is used with the global descriptors as predictors. Several class approaches are compared. Given our application, the best results are obtained with the Support Vector Machine with a combination of descriptors increasing the F1 by 0.031 with respect to the best descriptor alone.

Online Forecasting of Synchronous Time Series Based on Evolving Linear Models

This paper suggests evolving linear models as a powerful alternative for online forecasting of synchronous time series. First, using a priori knowledge of an expert, all known possible features are clustered in some categories so that each category includes homogeneous features. Then, a novel evolving correlation-based forward subset selection technique is used to determine relevant features from each category. Next, based on the selected features, an estimation of the output of the system is modeled through an evolving adaptive linear regression model. In evolving systems, selected features and their associated weights could vary over time based on new incoming data samples which contain new information. Finally, a soft combination of output estimations of all categories, in a new sense of Takagi–Sugeno fuzzy system, gives the prediction of the output of the system at each sampling time. The approach offers a certain new view at the enhancement of evolving forecasting models. The proposed approach embodies recursive learning and one-step-ahead incremental algorithms that progressively modify the model to ensure continuous learning, and self-organization of the model structure and its parameters. Two real-world problems, forecasting electricity load of the Electric Reliability Council of Texas region and stock price forecasting of technology sector of Standard & Poor’s 500 index, are provided to validate the developed method. Pros and cons of the proposed approach are comprehensively discussed and shown through simulation results and comparisons with other state-of-the-art techniques.

Using Serum Amino Acids to Predict Traumatic Brain Injury: A Systematic Approach to Utilize Multiple Biomarkers.

Traumatic brain injury (TBI) can cause biochemical and metabolomic alterations in the brain tissue and serum. These alterations can be used for diagnosis and prognosis of TBI. Here, the serum concentrations of seventeen amino acids (AA) were studied for their potential utility as biomarkers of TBI. Twenty-five female, 4-week-old piglets received diffuse (n = 13) or focal (n = 12) TBI. Blood samples were obtained both pre-injury and at either 24-h or 4-days post-TBI. To find a robust panel of biomarkers, the results of focal and diffuse TBIs were combined and multivariate logistic regression analysis, coupled with the best subset selection technique and repeated k-fold cross-validation method, was used to perform a thorough search of all possible subsets of AAs. The combination of serum glycine, taurine, and ornithine was optimal for TBI diagnosis, with 80% sensitivity and 86% overall prediction rate, and showed excellent TBI diagnostic performance, with 100% sensitivity and 78% overall prediction rate, on a separate validation dataset including four uninjured and five injured animals. We found that combinations of biomarkers outperformed any single biomarker. We propose this 3-AA serum biomarker panel to diagnose mild-to-moderate focal/diffuse TBI. The systematic approaches implemented herein can be used for combining parameters from various TBI assessments to develop/evaluate optimal multi-factorial diagnostic/prognostic TBI metrics.

Inter-Subset Hamming Distance Maximization for Enhancing the Physical Layer Security of Antenna Subset Modulation

In this article, a novel antenna subset selection technique for enhancing the Physical Layer Randomness (PLR) of Antenna Subset Modulation (ASM) has been proposed. Hamming Distance Optimized Antenna Subset Selection (HD-OASS) minimizes the correlation between the antenna subsets at transmitter by maximizing the hamming distance between the successively used antenna subsets. Unlike previously proposed Randomized Antenna Subset Selection (RASS) and Side-Lobe Level Optimized Antenna Subset Selection (SLL-OASS), in which antenna subsets are randomly selected from the codebook, HD-OASS chooses the antenna subsets having optimally maximized hamming distance. It is shown that SLL-OASS has rather unfavorable effects on encryption strength due to considerable reduction of codebook size. Furthermore, HD-OASS has been shown to outperform RASS and SLL-OASS in terms of encryption strength in the unwanted directions of eavesdropper.

Text Classification of Cornell Movie Data using Data Mining with Feature Selection

Text Classification is branch of text mining through which we can analyze the sentiment of the movie data. In this research paper we have applied different preprocessing techniques to reduce the features from cornell movie data set. We have also applied the Correlation-based feature subset selection and chi-square feature selection technique for gathering most valuable words of each category in text mining processes. The new cornell movie data set formed after applying the preprocessing steps and feature selection techniques. We have classified the cornell movie data as positive or negative using various classifiers like Support Vector Machine (SVM), Multilayer Perceptron (MLP), Naive Bayes (NB), Bays Net (BN) and Random Forest (RF) classifier. We have also compared the classification accuracy among classifiers and achieved better accuracy i. e. 87% in case of SVM classifier with reduced number of features. The suggested classifier can be useful in opinion of movie review, analysis of any blog and documents etc.