AndroClass: An Effective Method to Classify Android Applications by Applying Deep Neural Networks to Comprehensive Features

Android application (app) stores contain a huge number of apps, which are manually classified based on the apps’ descriptions into various categories. However, the predefined categories or apps descriptions are usually not very accurate to reflect the real functionalities of apps, thereby leading to misclassify the apps, which may cause serious security issues and unreliability problem in the app store. Therefore, the automatic app classification is an important demand to construct a secure, reliable, integrated, and easy to navigate app store. In this paper, we propose an effective method called AndroClass to automatically classify apps based on their real functionalities by using rich and comprehensive features representing the actual functionalities of the apps. AndroClass performs three steps of feature extraction, feature refinement, and classification. In the feature extraction step, we extract 14 various features for each app by utilizing a unified tool suite. In the feature refinement step, we apply Random Forest algorithm to refine the features. In the classification step, we combine refined features into a single one and AndroClass is equipped with K-Nearest Neighbor, Naive Bayes, Support Vector Machine, and Deep Neural Network to classify apps. On the contrary to the existing methods, all the utilized features in AndroClass are stable and clearly represent the actual functionalities of the app, AndroClass does not pose any issues to the user privacy, and our method can be applied to classify unreleased or newly released apps. The results of extensive experiments with two real-world datasets and a dataset constructed by human experts demonstrate the effectiveness of AndroClass where the classification accuracy of AndroClass with the latter dataset is 83.5%.

Download Full-text

A Comparison of the Analysis of Methods for Feature Extraction and Classification by Wavelet Transform in SSVEP BCIs

10.21203/rs.3.rs-82008/v1 ◽

2020 ◽

Author(s):

Hoda Heidari ◽

Zahra Einalou ◽

Mehrdad Dadgostar ◽

Hamidreza Hosseinzadeh

Keyword(s):

Feature Extraction ◽

Feature Selection ◽

Wavelet Transform ◽

Decision Tree ◽

Nearest Neighbor ◽

Support Vector ◽

K Nearest Neighbor ◽

Iir Filters ◽

Wide Range ◽

New Feature

Abstract Most of the studies in the field of Brain-Computer Interface (BCI) based on electroencephalography have a wide range of applications. Extracting Steady State Visual Evoked Potential (SSVEP) is regarded as one of the most useful tools in BCI systems. In this study, different methods such as feature extraction with different spectral methods (Shannon entropy, skewness, kurtosis, mean, variance) (bank of filters, narrow-bank IIR filters, and wavelet transform magnitude), feature selection performed by various methods (decision tree, principle component analysis (PCA), t-test, Wilcoxon, Receiver operating characteristic (ROC)), and classification step applying k nearest neighbor (k-NN), perceptron, support vector machines (SVM), Bayesian, multiple layer perceptron (MLP) were compared from the whole stream of signal processing. Through combining such methods, the effective overview of the study indicated the accuracy of classical methods. In addition, the present study relied on a rather new feature selection described by decision tree and PCA, which is used for the BCI-SSVEP systems. Finally, the obtained accuracies were calculated based on the four recorded frequencies representing four directions including right, left, up, and down.

Download Full-text

Analysis of classification methods suitable for band limited spatially filtered EEG signal applicable to non-invasive BCI

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.i8170.078919 ◽

2019 ◽

Vol 8 (9) ◽

pp. 1689-1694

Keyword(s):

Feature Extraction ◽

Nearest Neighbor ◽

Support Vector ◽

K Nearest Neighbor ◽

Common Spatial Pattern ◽

Linear Discriminant ◽

Feature Extraction Method ◽

Non Invasive ◽

Band Limited ◽

Ca Sensitivity

Electroencephalographic (EEG) signals are the preferred input for non-invasive Brain-Computer Interface (BCI). Efficient signal processing strategies, including feature extraction and classification, are required to distinguish the underlying task of BCI. This work proposes the optimized common spatial pattern(CSP) filtering technique as the feature extraction method for collecting the spatially spread variation of the signal. The bandpass filter (BPF) designed for this work assures the availability of event-related synchronized (ERS) and event-related desynchronized (ERD) signal as input to the spatial filter. This work takes consideration of the area-specific electrodes for feature formation. This work further proposes a comparative analysis of classifier algorithms for classification accuracy(CA), sensitivity and specificity and the considered algorithms are Support Vector Machine(SVM), Linear Discriminant Analysis(LDA), and K-Nearest Neighbor(KNN). Performance parameters considered are CA, sensitivity, and selectivity, which can judge the method not only for high CA but also inclining towards the particular class. Thus it will direct in the selection of appropriate classifier as well as tuning the classifier to get the balanced results. In this work, CA, the prior performance parameter is obtained to be 88.2% sensitivity of 94.2% and selectivity 82.2% for the cosine KNN classifier. SVM with linear kernel function also gives the comparable results, thus concluding that the robust classifiers perform well for all parameters in case of CSP for feature extraction.

Download Full-text

Biometric authenticator algorithm based on multiresolution analysis

Indonesian Journal of Electrical Engineering and Computer Science ◽

10.11591/ijeecs.v20.i3.pp1332-1341 ◽

2020 ◽

Vol 20 (3) ◽

pp. 1332

Author(s):

Soumia Kerrache ◽

Beladgham Mohammed ◽

Hamza Aymen ◽

Kadri Ibrahim

Keyword(s):

Feature Extraction ◽

Multiresolution Analysis ◽

Nearest Neighbor ◽

Curvelet Transform ◽

Principal Component ◽

Image Features ◽

Support Vector ◽

K Nearest Neighbor ◽

Feature Extraction Method ◽

Fusion Approach

Features extraction is an essential process in identifying person biometrics because the effectiveness of the system depends on it. Multiresolution Analysis success can be used in the system of a person’s identification and pattern recognition. In this paper, we present a feature extraction method for two-dimensional face and iris authentication. Our approach is a combination of principal component analysis (PCA) and curvelet transform as an improved fusion approach for feature extraction. The proposed fusion approach involves image denoising using 2D-Curvelet transform to achieve compact representations of curves singularities. This is followed by the application of PCA as a fusion rule to improve upon the spatial resolution. The limitations of the only PCA algorithm are a poor recognition speed and complex mathematical calculating load, to reduce these limitations, we are applying the curvelet transform. <br /> To assess the performance of the presented method, we have employed three classification techniques: Neural networks (NN), K-Nearest Neighbor (KNN) and Support Vector machines (SVM).<br />The results reveal that the extraction of image features is more efficient using Curvelet/PCA.

Download Full-text

Machine Learning Classification and Feature Extraction of Arrhythmic ECG Data

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.b3548.079220 ◽

2020 ◽

Vol 9 (2) ◽

pp. 6-12

Keyword(s):

Machine Learning ◽

Feature Extraction ◽

Nearest Neighbor ◽

Extraction Process ◽

Support Vector ◽

Ecg Signal ◽

Data Sets ◽

K Nearest Neighbor ◽

Machine Learning Classification ◽

Artificial Neural Network Ann

Electrocardiogram (ECG) is the analysis of the electrical movement of the heart over a period of time. The detailed information about the condition of the heart is measured by analyzing the ECG signal. Wavelet transform, fast Fourier transform are the different methods to disorganize cardiac disease. The paper elaborates the survey on ECG signal analysis and related study on arrhythmic and non arrhythmic data. Here we discuss the efficient feature extraction process for electrocardiogram, where based on position and priority six best P-QRS-T fragments are studied. This survey examines the the outcome of the system by using various Machine learning classification algorithms for feature extraction and analysis of ECG Signals. Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Artificial Neural Network (ANN) are the most important algorithms used here for this purpose. There are several publicly available data sets which are used for arrhythmia analysis and among them MIT-BIH ECG-ID database is mostly used. The drawbacks and limitations are also discussed here and from there future challenges and concluding remarks can be done.

Download Full-text

A comparison of the analysis of methods for feature extraction and classification by Wavelet transform in SSVEP BCIs

10.21203/rs.3.rs-82008/v2 ◽

2021 ◽

Author(s):

Hoda Heidari ◽

zahra einalou ◽

Mehrdad Dadgostar ◽

Hamidreza Hosseinzadeh

Keyword(s):

Feature Extraction ◽

Feature Selection ◽

Wavelet Transform ◽

Decision Tree ◽

Nearest Neighbor ◽

Support Vector ◽

K Nearest Neighbor ◽

Wide Range ◽

Vector Machines ◽

New Feature

Abstract Most of the studies in the field of Brain-Computer Interface (BCI) based on electroencephalography have a wide range of applications. Extracting Steady State Visual Evoked Potential (SSVEP) is regarded as one of the most useful tools in BCI systems. In this study, different methods which includes 1) feature extraction with different spectral methods (Shannon entropy, skewness, kurtosis, mean, variance) and wavelet transform magnitude, 2) feature selection performed by various methods (decision tree, principle component analysis (PCA), t-test, Wilcoxon, Receiver operating characteristic (ROC)), 3) classification step applying k nearest neighbor (k-NN), support vector machines (SVM), Bayesian, multiple layer perceptron (MLP) were compared from the whole stream of signal processing. Through combining such methods, the effective overview of the study indicated the accuracy of classical methods. In addition, the present study relied on a rather new feature selection described by decision tree and PCA, which is used for the BCI-SSVEP systems. Finally, the obtained accuracies were calculated based on the four recorded frequencies representing four directions including right, left, up, and down. The highest level of accuracy was obtained 91.39%.

Download Full-text

Hybrid Malware Classification Method Using Segmentation-Based Fractal Texture Analysis and Deep Convolution Neural Network Features

Applied Sciences ◽

10.3390/app10144966 ◽

2020 ◽

Vol 10 (14) ◽

pp. 4966 ◽

Cited By ~ 1

Author(s):

Maryam Nisa ◽

Jamal Hussain Shah ◽

Shansa Kanwal ◽

Mudassar Raza ◽

Muhammad Attique Khan ◽

...

Keyword(s):

Neural Network ◽

Feature Extraction ◽

Texture Analysis ◽

Data Augmentation ◽

Nearest Neighbor ◽

Feature Fusion ◽

Malicious Code ◽

Support Vector ◽

K Nearest Neighbor ◽

Affine Image

As the number of internet users increases so does the number of malicious attacks using malware. The detection of malicious code is becoming critical, and the existing approaches need to be improved. Here, we propose a feature fusion method to combine the features extracted from pre-trained AlexNet and Inception-v3 deep neural networks with features attained using segmentation-based fractal texture analysis (SFTA) of images representing the malware code. In this work, we use distinctive pre-trained models (AlexNet and Inception-V3) for feature extraction. The purpose of deep convolutional neural network (CNN) feature extraction from two models is to improve the malware classifier accuracy, because both models have characteristics and qualities to extract different features. This technique produces a fusion of features to build a multimodal representation of malicious code that can be used to classify the grayscale images, separating the malware into 25 malware classes. The features that are extracted from malware images are then classified using different variants of support vector machine (SVM), k-nearest neighbor (KNN), decision tree (DT), and other classifiers. To improve the classification results, we also adopted data augmentation based on affine image transforms. The presented method is evaluated on a Malimg malware image dataset, achieving an accuracy of 99.3%, which makes it the best among the competing approaches.

Download Full-text

Deep Fusion Feature Extraction for Caries Detection on Dental Panoramic Radiographs

Applied Sciences ◽

10.3390/app11052005 ◽

2021 ◽

Vol 11 (5) ◽

pp. 2005

Author(s):

Toan Huy Bui ◽

Kazuhiko Hamamoto ◽

May Phu Paing

Keyword(s):

Feature Extraction ◽

Nearest Neighbor ◽

Classification Model ◽

Caries Detection ◽

Support Vector ◽

K Nearest Neighbor ◽

Previous State ◽

Wide Scale ◽

Optimal Fusion ◽

Fusion Feature

Caries is the most well-known disease and relates to the oral health of billions of people around the world. Despite the importance and necessity of a well-designed detection method, studies in caries detection are still limited and show a restriction in performance. In this paper, we proposed a computer-aided diagnosis (CAD) method to detect caries among normal patients using dental radiographs. The proposed method mainly consists of two processes: feature extraction and classification. In the feature extraction phase, the chosen 2D tooth image was employed to extract deep activated features using a deep pre-trained model and geometric features using mathematic formulas. Both feature sets were then combined, called fusion feature, to complement each other defects. Then, the optimal fusion feature set was fed into well-known classification models such as support vector machine (SVM), k-nearest neighbor (KNN), decision tree (DT), Naïve Bayes (NB), and random forest (RF) to determine the best classification model that fit the fusion features set and perform the most preeminent result. The results show 91.70%, 90.43%, and 92.67% for accuracy, sensitivity, and specificity, respectively. The proposed method has outperformed the previous state-of-the-art and shows promising results when none of the measured factors is less than 90%; therefore, the method is promising for dentists and capable of wide-scale implementation caries detection in hospitals.

Download Full-text

Fault Diagnosis of Permanent Magnet DC Motors Based on Multi-Segment Feature Extraction

Sensors ◽

10.3390/s21227505 ◽

2021 ◽

Vol 21 (22) ◽

pp. 7505

Author(s):

Lixin Lu ◽

Weihao Wang

Keyword(s):

Feature Extraction ◽

Fault Diagnosis ◽

Permanent Magnet ◽

Nearest Neighbor ◽

Classification And Regression Tree ◽

Support Vector ◽

K Nearest Neighbor ◽

Single Segment ◽

Dc Motors ◽

Permanent Magnet Dc Motors

For permanent magnet DC motors (PMDCMs), the amplitude of the current signals gradually decreases after the motor starts. Only using the signal features of current in a single segment is not conducive to fault diagnosis for PMDCMs. In this work, multi-segment feature extraction is presented for improving the effect of fault diagnosis of PMDCMs. Additionally, a support vector machine (SVM), a classification and regression tree (CART), and the k-nearest neighbor algorithm (k-NN) are utilized for the construction of fault diagnosis models. The time domain features extracted from several successive segments of current signals make up a feature vector, which is adopted for fault diagnosis of PMDCMs. Experimental results show that multi-segment features have a better diagnostic effect than single-segment features; the average accuracy of fault diagnosis improves by 19.88%. This paper lays the foundation of fault diagnosis for PMDCMs through multi-segment feature extraction and provides a novel method for feature extraction.

Download Full-text

Efficient Classification of Horizontal And Vertical EOG Signals For Human Computer Interaction

10.21203/rs.3.rs-471385/v1 ◽

2021 ◽

Author(s):

Abdel-Gawad A. Abdel-Samei ◽

Ahmed S.Ali ◽

Fathi E. Abd El-Samie ◽

Ayman M.Brisha

Keyword(s):

Support Vector Machine ◽

Feature Extraction ◽

Eye Movements ◽

Human Computer Interaction ◽

Nearest Neighbor ◽

Human Interaction ◽

Support Vector ◽

K Nearest Neighbor ◽

Human Eye ◽

Computer Interaction

Abstract Human-computer interaction (HCI) using Electrooculography (EOG) has been a growing area of research in recent years. The HCI provides communication channels between the human and the external device. Today, EOG is one of the most important biomedical signals for measuring and analyzing the direction of eye movements. The EOG is used to produce both activities in vertical and horizontal directions of human eye movements. In this paper, different human eye movement tasks from vertical and horizontal directions are studied. The dataset of EOG signals were obtained from Electroencephalography (EEG) electrodes from 27 healthy people, 14 males and 13 females. This process resulted from two dipole signals, the vertical-EOG signals and the horizontal-EOG signals. These signals were filtered by band-pass at 0.5–5Hz. A total of 54 datasets from these 27 healthy individuals, each lasting 30 seconds, were given. The Bo-Hjorth parameter was implemented for feature extraction on the preprocessed EOG signals. For classification, Decision Tree (DT), K-Nearest Neighbor (KNN), Ensemble Classifier (EC), Kernel Naive Bayes (KNB) and Support Vector Machine (SVM)) were utilized. The obtained results reveal that the best classifiers on horizontal and vertical signals are the Support Vector Machine (SVM), the Cosine KNN and the Ensemble Subspace Discriminant with having 100% percentage accuracies. Through designing the proposed algorithm for feature extraction, the highest performance of classification can be obtained for rehabilitation purposes and other applications that help the handicapped to take decisions for better life quality, by providing possible human interaction with a computer.

Download Full-text

Recognition of Consumer Preference by Analysis and Classification EEG Signals

Frontiers in Human Neuroscience ◽

10.3389/fnhum.2020.604639 ◽

2021 ◽

Vol 14 ◽

Author(s):

Mashael Aldayel ◽

Mourad Ykhlef ◽

Abeer Al-Nafjan

Keyword(s):

Feature Extraction ◽

Nearest Neighbor ◽

Consumer Preference ◽

Machine Learning Algorithms ◽

Support Vector ◽

Discrete Wavelet ◽

K Nearest Neighbor ◽

Detection Systems ◽

Power Spectral ◽

Selection Of

Neuromarketing has gained attention to bridge the gap between conventional marketing studies and electroencephalography (EEG)-based brain-computer interface (BCI) research. It determines what customers actually want through preference prediction. The performance of EEG-based preference detection systems depends on a suitable selection of feature extraction techniques and machine learning algorithms. In this study, We examined preference detection of neuromarketing dataset using different feature combinations of EEG indices and different algorithms for feature extraction and classification. For EEG feature extraction, we employed discrete wavelet transform (DWT) and power spectral density (PSD), which were utilized to measure the EEG-based preference indices that enhance the accuracy of preference detection. Moreover, we compared deep learning with other traditional classifiers, such as k-nearest neighbor (KNN), support vector machine (SVM), and random forest (RF). We also studied the effect of preference indicators on the performance of classification algorithms. Through rigorous offline analysis, we investigated the computational intelligence for preference detection and classification. The performance of the proposed deep neural network (DNN) outperforms KNN and SVM in accuracy, precision, and recall; however, RF achieved results similar to those of the DNN for the same dataset.

Download Full-text