Feature Generation Algorithm Research Articles

Human–object interaction detection algorithm based on graph structure and improved cascade pyramid network

Aiming at the problem of insufficient use of human–object interaction (HOI) information and spatial location information in images, we propose a human–object​ interaction detection network based on graph structure and improved cascade pyramid. This network is composed of three branches, namely, graph branch, human–object branch and human pose branch. In graph branch, we propose a Graph-based Interactive Feature Generation Algorithm (GIFGA) to address the inadequate utilization of interaction information. GIFGA constructs an initial dense graph model by taking humans and objects as nodes and their interaction relationships as edges. Then, by traversing each node, the graph model is updated to generate the final interaction features. In human pose branch, we propose an Improved Cascade Pyramid Network (ICPN) to tackle the underutilization of spatial location information. ICPN extracts human pose features and maps both the object bounding boxes and extracted human pose maps onto the global feature map to capture the most discriminative interaction-related region features within the global context. Finally, the features from the three branches are fed into a Multi-Layer Perceptron (MLP) for fusion and then classified for recognition. Experimental results demonstrate that our network achieves mAP of 54.93% and 28.69% on the V-COCO and HICO-DET datasets, respectively.

Changes in Music Curriculum Education in Colleges and Universities Based on Cloud Classroom Assistive Technology

Abstract With the rapid development of science and technology, cloud classroom teaching has gradually become a trend. The cloud-based music teaching aid training system designed in this paper learns to map rhythmic and pitch features through music sequence generation technology. The onset detection algorithm of spectral flux detects the onset of the note. The Bezier curve is introduced into the melodic line feature generation algorithm to construct a polynomial function for interpolation between the first and the last endpoints, and the emotion is utilized to guide fractal music creation. Finally, the effectiveness of this auxiliary teaching system in college music teaching has been verified. The results show that the optimal Hamming distance of the system is 0.353 in the music melody generation experiment, which indicates that the system can quickly generate melody lines for students in music teaching and improve their learning efficiency. The system can significantly improve the rationality and efficiency of music skill training, which is worth promoting.

SFCWGAN-BiTCN with Sequential Features for Malware Detection

In the field of adversarial attacks, the generative adversarial network (GAN) has shown better performance. There have been few studies applying it to malware sample supplementation, due to the complexity of handling discrete data. More importantly, unbalanced malware family samples interfere with the analytical power of malware detection models and mislead malware classification. To address the problem of the impact of malware family imbalance on accuracy, a selection feature conditional Wasserstein generative adversarial network (SFCWGAN) and bidirectional temporal convolutional network (BiTCN) are proposed. First, we extract the features of malware Opcode and API sequences and use Word2Vec to represent features, emphasizing the semantic logic between API tuning and Opcode calling sequences. Second, the Spearman correlation coefficient and the whale optimization algorithm extreme gradient boosting (WOA-XGBoost) algorithm are combined to select features, filter out invalid features, and simplify structure. Finally, we propose a GAN-based sequence feature generation algorithm. Samples were generated using the conditional Wasserstein generative adversarial network (CWGAN) on the imbalanced malware family dataset, added to the trainset to supplement the samples, and trained on BiTCN. In comparison, in tests on the Kaggle and DataCon datasets, the model achieved detection accuracies of 99.56% and 96.93%, respectively, which were 0.18% and 2.98% higher than the models of other methods.

Automated detection of toxicophores and prediction of mutagenicity using PMCSFG algorithm.

Maximum common substructures (MCS) have received a lot of attention in the chemoinformatics community. They are typically used as a similarity measure between molecules, showing high predictive performance when used in classification tasks, while being easily explainable substructures. In the present work, we applied the Pairwise Maximum Common Subgraph Feature Generation (PMCSFG) algorithm to automatically detect toxicophores (structural alerts) and to compute fingerprints based on MCS. We present a comparison between our MCS-based fingerprints and 12 well-known chemical fingerprints when used as features in machine learning models. We provide an experimental evaluation and discuss the usefulness of the different methods on mutagenicity data. The features generated by the MCS method have a state-of-the-art performance when predicting mutagenicity, while they are more interpretable than the traditional chemical fingerprints.

SAMA: Spatially-Aware Model-Agnostic Machine Learning Framework for Geophysical Data

Geophysical data is a form of spatial data that suffers from various limitations when applying conventional machine learning algorithms and evaluation techniques. A key limitation facing models trained on geophysical data is their inability to generalize well when deployed to predict from new unseen data. We address the problem of inaccurate performance assessments of machine learning models, that stems from violating independence assumptions during the feature selection and evaluation phases of the learning process. Our proposed spatially-aware and model-agnostic (SAMA) framework provides a suite of spatially-aware feature generation, feature selection, and model validation algorithms that account for spatial characteristics of geophysical data. The framework is model agnostic, as it tackles data-related challenges that are not affected by the specific machine learning algorithm used to fit the data. To demonstrate the effectiveness of the proposed approach, it is applied to the water saturation mapping problem using a novel geophysical dataset to train a prediction model. The proposed spatially-aware models obtains an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R^{2}$ </tex-math></inline-formula> of 0.620, an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$RMSE$ </tex-math></inline-formula> of 0.220 for predicting water saturation for the Whole Region of the reservoir model box and an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R^{2}$ </tex-math></inline-formula> of 0.161, an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$RMSE$ </tex-math></inline-formula> of 0.263 for the Interwell Region. Extensive experiments on 5 additional unseen datasets show that the model maintains stable performance across different datasets, which demonstrates the ability of the SAMA framework to produce robust models that are transferable to new datasets.

Human-object Behavior Analysis Based on Interaction Feature Generation Algorithm

Human-object Behavior Analysis Based on Interaction Feature Generation Algorithm

Application of a New Feature Generation Algorithm in Intrusion Detection System

The intrusion detection system is designed to discover the abnormal behavior of the network system, but it has the problems of low detection accuracy, inability to perform fine detection, and huge time cost. Therefore, it is necessary to design a fast and accurate intrusion detection system. Therefore, this paper proposes a multigranularity feature generation + XGBOOST method to improve the intrusion detection system. First, we propose a multigranularity feature generation algorithm, which converts all features into discrete features with different numbers of categories. Different numbers of categories represent different granularities. We believe that the combination of multiple different granular features can achieve better accurate attack detection. Then, we use the proposed method to perform experimental verification on the four datasets of KDD99, NSL-KDD, UNSW_NB15, and CSE-CIC-IDS2018. For the KDD99 dataset, detection rates of 100%, 100%, and 99.43% can be achieved in the two-category, five-category, and multicategory tasks, respectively; for the NSL-KDD dataset, detection rates of 100%, 100%, and 90.84% can be achieved in the two-category, five-category, and multicategory tasks, respectively; for the UNSW_NB15 dataset, 100% detection rate can be achieved in the second and tenth categories; for the CSE-CIC-IDS2018 dataset, 100% detection rate can be achieved in the third classification. Experiments show that the proposed algorithm can achieve accurate and precise detection. Finally, we experiment with the multigranularity feature generation algorithm on multiple classifiers and multiple datasets to prove the generalization ability of the proposed feature generation algorithm and compare the proposed algorithm with the CFS algorithm to prove the efficiency of the algorithm.

SENMQTT-SET: An Intelligent Intrusion Detection in IoT-MQTT Networks Using Ensemble Multi Cascade Features

Recently, the number of Internet of Things (IoT) networks has been grown exponentially, which results in more data sharing between devices without appropriate security mechanisms. Since huge data management is involved, maintaining the time constraints between the devices in IoT networks is another significant issue. To address these issues, an intelligent intrusion detection system has been adapted to recognize or predict a cyber-attack using Elite Machine Learning algorithms (EML), and a lightweight protocol is used to manage the time-constrained issue. The experimental analysis of work is done on a testbed setup with the hardware and sensors connected using a lightweight Message Queue Telemetry Transport (MQTT) protocol. This comprises three parts: (i) collection of data with the help of a sensor for three different scenarios called SEN-MQTTSET; (ii) multi-context feature generation using an ensemble statistical multi-view cascade feature generation algorithm from the SEN-MQTTSET dataset; and (iii) evaluating the dataset using ML algorithms. The SEN-MQTTSET dataset has been created from the three scenarios, such as normal, attack on a subscriber, and attack on a broker. The multi-context feature is generated from the raw dataset using an ensemble statistical multi-view cascade feature generation algorithm. The EML is proposed to select the best model for intrusion detection among ML algorithms such as Logistic Regression, K-Nearest Neighbour, Random Forest, Naive Bias, Support Vector Machine, Gradient Boosting, and Decision Tree by the performance metrics such as accuracy, prediction time, F1-score, and others. The proposed dataset is validated and the accuracy is found to be above 99% for the considered system model. Different quality parameters have been carried out for legitimate and attack traffic features to calculate the delay between the IoT-MQTT network.

Performance Comparison of Feature Generation Algorithms for Mosaic Photoacoustic Microscopy

Mosaic imaging is a computer vision process that is used for merging multiple overlapping imaging patches into a wide-field-of-view image. To achieve a wide-field-of-view photoacoustic microscopy (PAM) image, the limitations of the scan range of PAM require a merging process, such as marking the location of patches or merging overlapping areas between adjacent images. By using the mosaic imaging process, PAM shows a larger field view of targets and preserves the quality of the spatial resolution. As an essential process in mosaic imaging, various feature generation methods have been used to estimate pairs of image locations. In this study, various feature generation algorithms were applied and analyzed using a high-resolution mouse ear PAM image dataset to achieve and optimize a mosaic imaging process for wide-field PAM imaging. We compared the performance of traditional and deep learning feature generation algorithms by estimating the processing time, the number of matches, good matching ratio, and matching efficiency. The analytic results indicate the successful implementation of wide-field PAM images, realized by applying suitable methods to the mosaic PAM imaging process.

Asynchronous event feature generation and tracking based on gradient descriptor for event cameras

Recently, the event camera has become a popular and promising vision sensor in the research of simultaneous localization and mapping and computer vision owing to its advantages: low latency, high dynamic range, and high temporal resolution. As a basic part of the feature-based SLAM system, the feature tracking method using event cameras is still an open question. In this article, we present a novel asynchronous event feature generation and tracking algorithm operating directly on event-streams to fully utilize the natural asynchronism of event cameras. The proposed algorithm consists of an event-corner detection unit, a descriptor construction unit, and an event feature tracking unit. The event-corner detection unit addresses a fast and asynchronous corner detector to extract event-corners from event-streams. For the descriptor construction unit, we propose a novel asynchronous gradient descriptor inspired by the scale-invariant feature transform descriptor, which helps to achieve quantitative measurement of similarity between event feature pairs. The construction of the gradient descriptor can be decomposed into three stages: speed-invariant time surface maintenance and extraction, principal orientation calculation, and descriptor generation. The event feature tracking unit combines the constructed gradient descriptor and an event feature matching method to achieve asynchronous feature tracking. We implement the proposed algorithm in C++ and evaluate it on a public event dataset. The experimental results show that our proposed method achieves improvement in terms of tracking accuracy and real-time performance when compared with the state-of-the-art asynchronous event-corner tracker and with no compromise on the feature tracking lifetime.

A similarity-based approach for the variant design of core-pulling mechanism in injection mold design

Core-pulling mechanism plays a vital role in producing plastic products with complex shapes. However, the design of core-pulling mechanism is still excessively depended on experience. As a result, low intelligent input often leads to high costs and long lead-time. In this paper, we propose a similarity-based approach for the variant design of core-pulling mechanism in injection mold design. An assemble model supporting variant design is established as the foundation of the case repository of core-pulling mechanism. A hierarchical assessment algorithm, which calculate the synthetic similarity of four characteristic parameters, viz., core-pulling type, core-pulling force, core-pulling distance, and shape description, is proposed to evaluate the similarity among undercut features. On the basis of the retrieved case with the highest similarity to the undercut feature of new plastic product, geometric model modification is performed by fine-tuning and head feature generation algorithm, and assemble tree modification is finished by the structural descriptions for assemble frame and assemble information of parts. Finally, case study is provided to validate the feasibility and effectiveness of the proposed approach.

Multiset Canonical Correlations Analysis of Bidimensional Intrinsic Mode Functions for Automatic Target Recognition of SAR Images.

A novel feature generation algorithm for the synthetic aperture radar image is designed in this study for automatic target recognition. As an adaptive 2D signal processing technique, bidimensional empirical mode decomposition is employed to generate multiscale bidimensional intrinsic mode functions from the original synthetic aperture radar images, which could better capture the broad spectral information and details of the target. And, the combination of the original image and decomposed bidimensional intrinsic mode functions could promisingly provide more discriminative information for correct target recognition. To reduce the high dimension of the original image as well as bidimensional intrinsic mode functions, multiset canonical correlations analysis is adopted to fuse them as a unified feature vector. The resultant feature vector highly reduces the high dimension while containing the inner correlations between the original image and decomposed bidimensional intrinsic mode functions, which could help improve the classification accuracy and efficiency. In the classification stage, the support vector machine is taken as the basic classifier to determine the target label of the test sample. In the experiments, the 10-class targets in the moving and stationary target acquisition and recognition dataset are classified to investigate the performance of the proposed method. Several operating conditions and reference methods are setup for comprehensive evaluation.

Variable step dynamic threshold local binary pattern for classification of atrial fibrillation

In this paper, we proposed new methods for feature extraction in machine learning-based classification of atrial fibrillation from ECG signal. Our proposed methods improved conventional 1-dimensional local binary pattern method in two ways. First, we proposed a dynamic threshold LBP code generation method for use with 1-dimensional signals, enabling the generated LBP codes to have a more detailed representation of the signal morphological pattern. Second, we introduced a variable step value into the LBP code generation algorithm to better cope with a high sampling frequency input signal without a downsampling process. The proposed methods do not employ computationally expensive processes such as filtering, wavelet transform, up/downsampling, or beat detection, and can be implemented using only simple addition, division, and compare operations. Combining these two approaches, our proposed variable step dynamic threshold local binary pattern method achieved 99.11% sensitivity and 99.29% specificity when used as a feature generation algorithm in support vector machine classification of atrial fibrillation from MIT-BIH Atrial Fibrillation Database dataset. When applied on signals from MIT-BIH Arrhythmia Database, our proposed method achieved similarly good 99.38% sensitivity and 98.97% specificity. Our proposed methods achieved one of the best results among published works in atrial fibrillation classification using the same dataset while using less computationally expensive calculations, without significant performance degradation when applied on signals from multiple databases with different sampling frequencies.

Dependence Representation Learning with Convolutional Neural Networks and 2D Histograms

Researchers frequently use visualizations such as scatter plots when trying to understand how random variables are related to each other, because a single image represents numerous pieces of information. Dependency measures have been widely used to automatically detect dependencies, but these measures only take into account a few types of data, such as the strength and direction of the dependency. Based on advances in the applications of deep learning to vision, we believe that convolutional neural networks (CNNs) can come to understand dependencies by analyzing visualizations, as humans do. In this paper, we propose a method that uses CNNs to extract dependency representations from 2D histograms. We carried out three sorts of experiments and found that CNNs can learn from visual representations. In the first experiment, we used a synthetic dataset to show that CNNs can perfectly classify eight types of dependency. Then, we showed that CNNs can predict correlations based on 2D histograms of real datasets and visualize the learned dependency representation space. Finally, we applied our method and demonstrated that it performs better than the AutoLearn feature generation algorithm in terms of average classification accuracy, while generating half as many features.

A hybrid sentiment analysis method for Turkish

This paper presents a hybrid methodology for Turkish sentiment analysis, which combines the lexicon-based and machine learning (ML)-based approaches. On the lexicon-based side, we use a sentiment dictionary that is extended with a synonyms lexicon. Besides this, we tackle the classification problem with three supervised classifiers, naive Bayes, support vector machines, and J48, on the ML side. Our hybrid methodology combines these two approaches by generating a new lexicon-based value according to our feature generation algorithm and feeds it as one of the features to machine learning classifiers. Despite the linguistic challenges caused by the morphological structure of Turkish, the experimental results show that it improves the accuracy by 7 % on average.

Multi-embed nonlinear scale-space for image trust root generation

An image trust root is a special type of soft trust root for trusted computing. However, image trust root generation is difficult, as it needs a corresponding stable logic feature generation model and algorithm for dynamical and sustained authentication. This paper proposes a basic function of constructing new scale-spaces with deep detecting ability and high stability for image features aimed at image root generation. According to the heat distribution and spreading principle of various kinds of infinitesimal heat sources in the space medium, a multi-embed nonlinear diffusion equation that corresponds to the multi-embed nonlinear scale-space is proposed, a HARRIS-HESSIAN scale-space evaluation operator that aims at the structure acceleration characteristics of a local region and can make use of image pixels' relative spreading movement principle was constructed, then a single-parameter global symmetric proportion (SPGSP) operator was also constructed. An authentication test with 3000 to 5000 cloud entities shows the new scale-space can work well and is stable, when the whole cloud has 5%–50% behavior with un-trusted entities. Consequently, it can be used as the corresponding stable logic feature generation model and algorithm for all kinds of images, and logic relationships among image features for trust roots.

A clustering technique for partial discharge and noise sources identification in power cables by means of waveform parameters

On-line partial discharge (PD) measurements have become a common technique for assessing the insulation condition of installed high voltage (HV) insulated cables. When on-line tests are performed in noisy environments, or when more than one source of pulse-shaped signals are present in a cable system, it is difficult to perform accurate diagnoses. In these cases, an adequate selection of the non-conventional measuring technique and the implementation of effective signal processing tools are essential for a correct evaluation of the insulation degradation. Once a specific noise rejection filter is applied, many signals can be identified as potential PD pulses, therefore, a classification tool to discriminate the PD sources involved is required. This paper proposes an efficient method for the classification of PD signals and pulse-type noise interferences measured in power cables with HFCT sensors. By using a signal feature generation algorithm, representative parameters associated to the waveform of each pulse acquired are calculated so that they can be separated in different clusters. The efficiency of the clustering technique proposed is demonstrated through an example with three different PD sources and several pulse-shaped interferences measured simultaneously in a cable system with a high frequency current transformer (HFCT).

Classification and Regression Analysis of the Prognostic Breast Cancer using Generation Optimizing Algorithms

Breast cancer is one of the main causes of female fatality all over the world and is the major field of research since quite a long time with lesser improvement than expected. Many institutions and organizations are working in this field to lead to a possible solution of the problem or to lead to more understanding of the problem. Many previous researches were studied for better understanding of the problem and the work done already to remove redundancy and contribute to the field, Wisconsin-Madison prognostic Breast cancer (WPBC) data set from the UCI machine learning repository was used for training of 198 individual cases by selecting best features out of 34 predictors. Feature selection algorithms were used with machine learning algorithms for feature reduction and for better classification. Different feature selection and generation algorithms were used to improve the accuracy of classification. Many improvements in accuracies were found out by using different approaches than the earlier studies conducted in the same field. The Naive Bayes and Logistic Regression algorithms showed 8.28-12.32% and 0.82-1.52% accuracy via 10 fold cross validation analysis improvement accordingly by using different feature selection and generation algorithms with these classifiers and gave better result than the best results known for these classification algorithms.

A Comprehensive Study of Features and Algorithms for URL-Based Topic Classification

Given only the URL of a Web page, can we identify its topic? We study this problem in detail by exploring a large number of different feature sets and algorithms on several datasets. We also show that the inherent overlap between topics and the sparsity of the information in URLs makes this a very challenging problem. Web page classification without a page’s content is desirable when the content is not available at all, when a classification is needed before obtaining the content, or when classification speed is of utmost importance. For our experiments we used five different corpora comprising a total of about 3 million (URL, classification) pairs. We evaluated several techniques for feature generation and classification algorithms. The individual binary classifiers were then combined via boosting into metabinary classifiers. We achieve typical F-measure values between 80 and 85, and a typical precision of around 86. The precision can be pushed further over 90 while maintaining a typical level of recall between 30 and 40.

Features generated for computational splice-site prediction correspond to functional elements.

BackgroundAccurate selection of splice sites during the splicing of precursors to messenger RNA requires both relatively well-characterized signals at the splice sites and auxiliary signals in the adjacent exons and introns. We previously described a feature generation algorithm (FGA) that is capable of achieving high classification accuracy on human 3' splice sites. In this paper, we extend the splice-site prediction to 5' splice sites and explore the generated features for biologically meaningful splicing signals.ResultsWe present examples from the observed features that correspond to known signals, both core signals (including the branch site and pyrimidine tract) and auxiliary signals (including GGG triplets and exon splicing enhancers). We present evidence that features identified by FGA include splicing signals not found by other methods.ConclusionOur generated features capture known biological signals in the expected sequence interval flanking splice sites. The method can be easily applied to other species and to similar classification problems, such as tissue-specific regulatory elements, polyadenylation sites, promoters, etc.