Feature Weight Vector Research Articles

Multi-objective evolutionary algorithm based on transfer learning and neural networks: Dual operator feature fusion and weight vector adaptation

In multi-objective evolutionary algorithms, one of the focal points is finding a balance between diversity and convergence. In decomposition-based algorithms, the role of weight vectors is crucial. Despite numerous studies dedicated to these aspects, there is a scarcity of utilizing transfer learning algorithms for dual-operator feature fusion and employing neural networks for accurate partitioning of the objective space population. To address the aforementioned issues, this paper proposes the following improvements: (1) Implementing a Balanced Distribution Adaptation (BDA) transfer learning algorithm to achieve dual-operator feature fusion, resulting in a transfer population guiding the adaptive adjustment of weight vectors. (2) Integrating the BDA algorithm with multi-objective algorithms requires labeling the data, a challenge in the multi-objective evolutionary algorithm. To tackle this issue, non-dominated sorting is introduced as a bridge connecting the BDA and multi-objective evolutionary algorithms. This serves as a method to combine the advantages of decomposition-based and Pareto-dominance principle-based multi-objective algorithms. (3) To overcome the impact of traditional Euclidean distance on population sparsity, a neural network is employed to determine the population's distribution in the objective space accurately. This ensures the precise identification of individuals to be removed from the current population and the areas where additions are needed. In order to fully validate the effectiveness of the proposed algorithm, four different sets of experiments are conducted in the experimental section, where three sets of benchmarking problems are compared to a variety of algorithms that have received much attention in recent years, as well as ablation experiments.

Hybrid Deep Transfer Learning and Feature Fusion Architecture for Diabetic Retinopathy Classification and Severity Grading

Diabetic Retinopathy (DR) is the leading cause of blindness among individuals with diabetes. Automating the diagnosis of DR has the potential to greatly benefit patients by enabling early detection and intervention, thus reducing the risk of blindness. The primary objective of this research is to develop a robust approach for the classification of DR and to analyze its severity grading. By achieving this, we aim to provide an effective tool for accurate diagnosis and assessment of DR, contributing to improved patient care and outcomes. The current literature review analysis reported the importance of deep learning in computer vision based applications. Moreover, plenty of pre-trained models are also present which can be used for classification tasks. Therefore, we present a hybrid DL classification approach by combining Inception V3, VGG-19 and ResNet 50. The proposed architecture uses transfer learning, and feature fusion model to produce the weighted feature vector which is used for classification analysis. The proposed approach is experimented on publicly available datasets APTOS-2019 and Messidor. The performance is measured in terms of accuracy, precision, recall and F1-score.

Discriminative label correlation based robust structure learning for multi-label feature selection

Feature selection is a key technique to tackle the curse of dimensionality in multi-label learning. Lots of embedded multi-label feature selection methods have been developed. However, they face challenges in identifying and excluding redundant features. To address these issues, this paper proposes a multi-label feature selection method that combines robust structural learning and discriminative label regularization. The proposed method starts from the feature space rather than data space, motivated by the principle that redundant features have high similarity or strong correlation. To exclude redundant features, a regularization on the feature selection matrix is designed by combining ℓ2,1-norm penalty with inner products of feature weight vectors. This regularization can help to learn a robust structure in the feature selection matrix. Meanwhile, both of the similarity and dissimilarity of labels of instances are involved in exploring discriminative label correlations. Extensive experiments verified the effectiveness of the proposed model for feature selection.

Principal Component Approximation Network for Image Compression

In this work, we propose a novel principal component approximation network (PCANet) for image compression. The proposed network is based on the assumption that a set of images can be decomposed into several shared feature matrices, and an image can be reconstructed by the weighted sum of these matrices. The proposed PCANet is specifically devised to learn and approximate these feature matrices and weight vectors, which are used to encode images for compression. Unlike previous deep learning-based methods, a distinctive aspect of our approach is its consideration of network size in the bit-rate computation. Despite this inclusion, our proposed method yields promising results. Through extensive experiments conducted on standard datasets, we demonstrate the effectiveness of our approach in comparison to state-of-the-art techniques. To the best of our knowledge, this is the first machine learning approach that includes the size of networks during bitrate computation in image compression.

Classification method of traditional Chinese medicine compound decoction duration based on multi-dimensional feature weighted fusion

This paper extends a text classification method utilizing natural language processing (NLP) into the field of traditional Chinese medicine (TCM) compound decoction to effectively and scientifically extend the TCM compound decoction duration. Specifically, a TCM compound decoction duration classification named TCM-TextCNN is proposed to fuse multi-dimensional herb features and improve TextCNN. Indeed, first, we utilize word vector technology to construct feature vectors of herb names and medicinal parts, aiming to describe the herb characteristics comprehensively. Second, considering the impact of different herb features on the decoction duration, we use an improved Term Frequency-Inverse Word Frequency (TF-IWF) algorithm to weigh the feature vectors of herb names and medicinal parts. These weighted feature vectors are then concatenated to obtain a multi-dimensional herb feature vector, allowing for a more comprehensive representation. Finally, the feature vector is input into the improved TextCNN, which uses k-max pooling to reduce information loss rather than max pooling. Three fully connected layers are added to generate higher-level feature representations, followed by softmax to obtain the final results. Experimental results on a dataset of TCM compound decoction duration demonstrate that TCM-TextCNN improves accuracy, recall, and F1 score by 5.31%, 5.63%, and 5.22%, respectively, compared to methods solely rely on herb name features, thereby confirming our method’s effectiveness in classifying TCM compound decoction duration.

Intelligent Fault Diagnosis of Hydraulic Multi-Way Valve Using the Improved SECNN-GRU Method with mRMR Feature Selection

Hydraulic multi-way valves as core components are widely applied in engineering machinery, mining machinery, and metallurgical industries. Due to the harsh working environment, faults in hydraulic multi-way valves are prone to occur, and the faults that occur are hidden. Moreover, hydraulic multi-way valves are expensive, and multiple experiments are difficult to replicate to obtain true fault data. Therefore, it is not easy to achieve fault diagnosis of hydraulic multi-way valves. To address this problem, an effective intelligent fault diagnosis method is proposed using an improved Squeeze-Excitation Convolution Neural Network and Gated Recurrent Unit (SECNN-GRU). The effectiveness of the method is verified by designing a simulation model for a hydraulic multi-way valve to generate fault data, as well as the actual data obtained by establishing an experimental platform for a directional valve. In this method, shallow statistical features are first extracted from data containing fault information, and then fault features with high correlation with fault types are selected using the Maximum Relevance Minimum Redundancy algorithm (mRMR). Next, spatial dimension features are extracted through CNN. By adding the Squeeze-Excitation Block, different weights are assigned to features to obtain weighted feature vectors. Finally, the time-dimension features of the weighted feature vectors are extracted and fused through GRU, and the fused features are classified using a classifier. The fault data obtained from the simulation model verifies that the average diagnostic accuracy of this method can reach 98.94%. The average accuracy of this method can reach 92.10% (A1 sensor as an example) through experimental data validation of the directional valve. Compared with other intelligent diagnostic algorithms, the proposed method has better stationarity and higher diagnostic accuracy, providing a feasible solution for fault diagnosis of the hydraulic multi-way valve.

Attentive Siamese Networks for Automatic Modulation Classification Based on Multitiming Constellation Diagrams.

Automatic modulation classification (AMC) is an essential part in a cognitive radio receiver. Benefited from the discriminative constellation characteristics among most modulations, AMC methods based on constellation diagrams usually achieve pleasant performance. However, in noncooperation communication systems, constellation diagrams expressing modulations explicitly are difficult to obtain via blind symbol timing synchronization, especially in complicated wireless channels. Therefore, this article proposes a novel constellation diagram-based AMC architecture called attentive Siamese networks (ASNs) by considering multitiming constellation diagrams (MCDs) and selecting the proper symbol timings at the feature level, which is a more robust way than the conventional signal-level symbol timing synchronization. In detail, convolutional neural networks sharing the same parameters first extract deep feature vectors for MCDs. Then, an attention inference module weights all the deep feature vectors. Finally, AMC is realized based on the weighted feature vectors. Moreover, the ASN architecture can be trained end-to-end. Comparing with the state-of-the-art methods that take diverse representations of received baseband signals as input, experimental results based on the RadioML 2018.01A dataset and non-Gaussian noise dataset demonstrate that ASN achieves a remarkable improvement, whose classification accuracy goes over 99% when the signal-to-noise ratio (SNR) > 10 dB.

An object detection approach with residual feature fusion and second‐order term attention mechanism

AbstractAutomatically detecting and locating remote occlusion small objects from the images of complex traffic environments is a valuable and challenging research. Since the boundary box location is not sufficiently accurate and it is difficult to distinguish overlapping and occluded objects, the authors propose a network model with a second‐order term attention mechanism and occlusion loss. First, the backbone network is built on CSPDarkNet53. Then a method is designed for the feature extraction network based on an item‐wise attention mechanism, which uses the filtered weighted feature vector to replace the original residual fusion and adds a second‐order term to reduce the information loss in the process of fusion and accelerate the convergence of the model. Finally, an objected occlusion regression loss function is studied to reduce the problems of missed detections caused by dense objects. Sufficient experimental results demonstrate that the authors’ method achieved state‐of‐the‐art performance without reducing the detection speed. The mAP@.5 of the method is 85.8% on the Foggy_cityscapes dataset and the mAP@.5 of the method is 97.8% on the KITTI dataset.

Decentralized adaptively weighted stacked autoencoder-based incipient fault detection for nonlinear industrial processes

Modern industrial processes often exhibit large-scale and nonlinear characteristics. Incipient fault detection for industrial processes is a big challenge because of the faint fault signature. To improve the performance of incipient fault detection for large-scale nonlinear industrial processes, a decentralized adaptively weighted stacked autoencoder (DAWSAE) -based fault detection method is proposed. First, the industrial process is divided into several sub-blocks and local adaptively weighted stacked autoencoder (AWSAE) is established for each sub-block to mine local information and obtain local adaptively weighted feature vectors and residual vectors. Second, the global AWSAE is established for the whole process to mine global information and obtain global adaptively weighted feature vectors and residual vectors. Finally, local statistics and global statistics are constructed based on local and global adaptively weighted feature vectors and residual vectors to detect the sub-blocks and the whole process, respectively. The advantages of proposed method are verified by a numerical example and Tennessee Eastman process (TEP).

G2Grad-CAMRL: An Object Detection and Interpretation Model Based on Gradient-Weighted Class Activation Mapping and Reinforcement Learning in Remote Sensing Images

Remote sensing images contain important information such as airports, ports, and ships. By extracting remote sensing image features and learning the mapping relationship between image features and text semantic features, the interpretation and description of remote sensing image content can be realized, which has a wide range of application value in military and civil fields such as national defense security, land monitoring, urban planning, disaster mitigation and so on. Aiming at the complex background of remote sensing images and the lack of interpretability of existing target detection models, and the problems in feature extraction between different network structures, different layers, and the accuracy of target classification, we propose an object detection and interpretation model based on Gradient weighted class activation mapping and reinforcement learning. Firstly, ResNet is used as the main backbone network to extract the features of remote sensing images and generate feature graphs. Then, we add the global average pooling layer to obtain the corresponding feature weight vector of the feature graph. The weighted vectors are superimposed to output class activation maps. The reinforcement learning method is used to optimize the generated region generation network. At the same time, we improve the reward function of reinforcement learning to improve the effect of the region generation network. Finally, network dissecting analysis is used to obtain the interpretable semantic concept in the model. Through experiments, the average accuracy is more than 85%. Experimental results in the public remote sensing image description dataset show that the proposed method has high detection accuracy and good description performance for remote sensing images in complex environments.

Design and Development of New Interactive Chatbot System for Mobile Service Providers through Heuristic-Based Ensemble Learning

In recent times, the human agents for chat services are used to be more in the communication system among human users according to artificial intelligence. These chatbots seem to be cost-efficient and time-saving paradigm, and so it is widely developed for usage. Still, it suffers from satisfying the customer requirements, which reduces the users being less liable to conform to desires through the chatbot. The designed interactive system has to process the queries and gets a suitable reply in the chatbot. The data collection related to diverse queries is gathered. Then, the pre-processing is carried out. Further, the word to vector is conducted for extracting the features. Then, the weighted feature vector is attained through optimizing the weight by hybrid optimization named Hybrid Water Wave Barnacle Mating Optimization (HWW-BMO). Finally, detection is performed through Heuristic-based Extreme learning + Long Short-Term Memory (H-ExtLSTM) for predicting the intent related to the queries asked by mobile service providers, thus can extract the answers based on word similarity. Here, the parameters of ELM and LSTM are optimized by the HWW-BMO algorithm. Experiments show that the proposed joint models have achieved competitive results in different datasets in multiple scenarios compared with other models.

Research on Agricultural Economic Early Warning Based on Genetic Algorithm and SVM

Agriculture is unique in that the industry is subject to a certain level of uncertainty due to seasonal and other factors, and its risk level is significantly higher than that of other industries. Therefore, it is necessary to establish an appropriate financial early warning model to predict, analyze, and control its financial risks. The article uses a genetic algorithm and support vector machine-based economic forecasting model for agribusinesses to adapt its own pollutant weights in a practical application environment to improve the forecasting accuracy. This model first uses a genetic algorithm to train a feature weight vector of current data so that the weights are adapted to the current prediction problem and then uses this feature weight vector to apply to SVM model training. It is found that 62.79% of the listed agricultural companies are in warning status, and their development is not optimistic. The article provides new ideas for the subsequent research on financial warning methods and also expands the boundaries of theoretical research for the research system of financial warning problems and enriches the experience and evidence of practical research.

A Hybrid Feature Selection and Ensemble Approach to Identify Depressed Users in Online Social Media.

Depression has become one of the most common mental illnesses, and the widespread use of social media provides new ideas for detecting various mental illnesses. The purpose of this study is to use machine learning technology to detect users of depressive patients based on user-shared content and posting behaviors in social media. At present, the existing research mostly uses a single detection method, and the unbalanced class distribution often leads to a low recognition rate. In addition, a large number of irrelevant or redundant features in high-dimensional data sets interfere with the accuracy of recognition. To solve this problem, this paper proposes a hybrid feature selection and stacking ensemble strategy for depression user detection. First, recursive elimination method and extremely randomized trees method are used to calculate feature importance and mutual information value, calculate feature weight vector, and select the optimal feature subset according to the feature weight. Second, naive bayes, k-nearest neighbor, regularized logistic regression and support vector machine are used as base learners, and a simple logistic regression algorithm is used as a combination strategy to build a stacking model. Experimental results show that compared with other machine learning algorithms, the proposed hybrid method, which integrates feature selection and ensemble, has a higher accuracy of 90.27% in identifying online patients. We believe this study will help develop new methods to identify depressed people in social networks, providing guidance for future research.

Attention-based deep learning networks for identification of human gait using radar micro-Doppler spectrograms

AbstractIdentification of human individuals within a group of 39 persons using micro-Doppler (μ-D) features has been investigated. Deep convolutional neural networks with two different training procedures have been used to perform classification. Visualization of the inner network layers revealed the sections of the input image most relevant when determining the class label of the target. A convolutional block attention module is added to provide a weighted feature vector in the channel and feature dimension, highlighting the relevant μ-D feature-filled areas in the image and improving classification performance.

Sequence Based DNA-Binding Protein Prediction

Protein and DNA have vital role in our biological processes. For accurately predicting DNA binding protein, develop a new sequence based prediction method from the protein sequence. Sequence based method only considers the protein sequence information as input. For accurately predicting DBP, first develop a reliable benchmark data set from the protein data bank. Second, using Amino Acid Composition (AAC), Position Specific Scoring Matrix (PSSM), Predicted Solvent Accessibility (PSA), and Predicted Probabilities of DNA-Binding Sites (PDBS) to produce four specific protein sequence baselines. Using a differential evolution algorithm, weights of the properties are taught. Based on those attained properties, merge the characteristics with weights to create an original super feature. And tensor-flow is used to paralyze the weights. A suitable feature selection algorithm of tensor flow’s binary classifier is used to extract the excellent subset from weighted feature vector. The training sample set is obtained in the training process, after generating final features. The classification is learned through the support vector machine and the tensor flow. And the output is measured using a tensor surface. The choice is done on the basis of threshold of likelihood and protein with above-threshold chance is considered to be DBP and others are non-DBP.

Improved detection performance in blood cell count by an attention-guided deep learning method

Blood cell count plays an important role in the field of clinical medical diagnosis. To effectively automate the counting of blood cells, recently, the deep-learning-based detection method represented by the YOLO has been proposed and used successfully. Nevertheless, the YOLO detection method has difficulties in insufficient positioning of the bounding boxes and in distinguishing overlapping objects. To overcome the limitations, we propose a new deep-learning-based method, termed Attention-YOLO, which is achieved by adding the channel attention mechanism and the spatial attention mechanism to the feature extraction network. By using the filtered and weighted feature vector to replace the original feature vector for residual fusion, Attention-YOLO can help the network to improve the detection accuracy. The experimental results indicate that compared to the standard YOLO network, the Attention-YOLO can achieve a better detection performance in blood cell count without introducing too many additional parameters, where the recognition accuracy of cells (RBCs, WBCs, and platelets) has an improvement of 6.70%, 2.13%, and 10.44%, respectively, and the mean Average Precision (mAP) has an improvement of 7.10%.

Multifeature Metric Learning Based on Enhanced Equidistance Embedding for Electroencephalogram Recognition of Epilepsy

Mobile edge computing (MEC) has the ability of pattern recognition and intelligent processing of real‐time data. Electroencephalogram (EEG) is a very important tool in the study of epilepsy. It provides rich information that can not be provided by other physiological methods. In the automatic classification of EEG signals by intelligent algorithms, feature extraction and the establishment of classifiers are both very important steps. Different feature extraction methods, such as time domain, frequency domain, and nonlinear dynamic feature methods, contain independent and diverse specific information. Using multiple forms of features at the same time can improve the accuracy of epilepsy recognition. In this paper, we apply metric learning to epileptic EEG signal recognition. Inspired by the equidistance constrained metric learning algorithm, we propose multifeature metric learning based on enhanced equidistance embedding (MMLE3) for EEG recognition of epilepsy. The MMLE3 algorithm makes use of various forms of EEG features, and the feature weights are adaptively weighted. It is a big advantage that the feature weight vector can be adjusted adaptively, without manual adjustment. The MMLE3 algorithm maximizes the distance between the samples constrained by the cannot‐link, and the samples of different classes are transformed into equidistant; meanwhile, MMLE3 minimizes the distance between the data constrained by the must‐link, and the samples of the same class are compressed to one point. Under the premise that the various feature classification tasks are consistent, MMLE3 can fully extract the associated and complementary information hidden between the features. The experimental results on the CHB‐MIT dataset verify that the MMLE3 algorithm has good generalization performance.

Improving the similarity search between images of agricultural products: An approach based on fuzzy rough theory

Many methods have been presented in recent years for identifying the quality of agricultural products using machine vision that due to the huge amount of redundant information and noisy data of images of products, the retrieval accuracy and speed of such methods were not much acceptable. All of them try to provide approaches to extract efficient features and determine optimal methods to measure similarity between images. One of the basic problems of these methods is determination of desirable features of the user as well as using an appropriate similarity measure. This study tries to recognize the importance of each feature according to user’s opinion in every feedback stage through using weighted feature vector, rough theory and fuzzy logic for identifying important features and finding a higher accuracy in retrieval result. The proposed method is compared with fuzzy color histogram, combined approach and fuzzy neighborhood entropy characterized by color location. The simulation results indicate that the proposed method has higher applicability in image marketing compared to the existing methods.

Optimal feature extraction and classification-oriented medical insurance prediction model: machine learning integrated with the internet of things

This paper plans to develop an effective machine learning system integrated with the Internet of Things (IoT) to predict the health insurance amount. IoT in healthcare enables interoperability, machine-to-machine communication, information exchange, and data movement that make healthcare service delivery effective. The model includes three phases (a) Feature Extraction, and (b) Weighted Feature Extraction, and (c) Prediction. The feature extraction process computes two statistical measures: First Order Statistics like mean, median, standard deviation, the maximum value of entire data, and minimum value of entire data, and Second-Order Statistics like Kurtosis, skewness, correlation, and entropy. The prediction process deploys a renowned machine learning algorithm called Neural Network (NN). As the main contribution, the weighted feature vector is developed here, where the weight optimally tuned by Modified Whale Optimization Algorithm (WOA). Also, the contribution relies on NN, where the training algorithm replaced with the same modified WOA for weight update. The modified WOA developed here is termed as Fitness dependent Randomized Whale Optimization Algorithm (FR-WOA). At last, the valuable experimental analysis using three datasets confirms the efficient performance of the suggested model.

Optimal weighted feature vector and deep belief network for medical data classification

Medical data classification is the process of transforming descriptions of medical diagnoses and procedures into universal medical code numbers. The diagnoses and procedures are usually taken from a variety of sources within the healthcare record, such as the transcription of the physician’s notes, laboratory results, radiologic results and other sources. However, there exist many frequency distribution problems in these domains. Hence, this paper intends to develop an advanced and precise medical data classification approach for diabetes and breast cancer dataset. With the knowledge of the features and challenges persisting with the state-of-the-art classification methods, deep learning-based medical data classification methodology is proposed here. It is well known that deep learning networks learn directly from the data. In this paper, the medical data is dimensionally reduced using Principle Component Analysis (PCA). The dimensionally reduced data are transformed by multiplying by a weighting factor, which is optimized using Whale Optimization Algorithm (WOA), to obtain the maximum distance between the features. As a result, the data are transformed into a label-distinguishable plane under which the Deep Belief Network (DBN) is adopted to perform the deep learning process, and the data classification is performed. Further, the proposed WOA-based DBN (WOADBN) method is compared with the Neural Network (NN), DBN, Generic Algorithm-based NN (GANN), GADBN, Particle Swarm Optimization (PSONN), PSO-based DBN (PSODBN), WOA-based NN (WOANN) techniques and the results are obtained, which shows the superiority of proposed algorithm over conventional methods.