Final Feature Vector Research Articles

Deep hybrid approach with sequential feature extraction and classification for robust malware detection

Malware attacks have escalated significantly with an increase in the number of internet users and connected devices. With the increasingly different types of malware released by hackers, designing new and competitive techniques to detect advanced malware is essential. In the proposed research, we have developed a multi-level feature extraction technique using deep learning architectures and a classification model to classify malware families. The essential features from the malware images are extracted using the Gated Recurrent Unit in the first step, which are further fed to a Convolutional Neural Network model for extracting the final feature vector. The multi-level feature selection is followed by classification into various malware families using Cost-sensitive Boot Strapped Weighted Random Forest (CSBW-RF). The proposed approach gave promising results of 99.58 % accuracy in distinguishing the 25 different malware families on the Mallmg dataset. This hybrid model gave significantly better performance scores for classifying visually similar malware families. The generalizability of the proposed model is benchmarked with the popular Microsoft Big 2015 dataset and has achieved comparatively higher performance scores than many existing models. This benchmarking demonstrates the robustness and scalability of our approach. The use of cost-sensitive learning and bootstrapping techniques also contributed to the model’s ability to generalize well to new and unseen data. These enhancements ensure that our model can be effectively applied in diverse real-world scenarios, maintaining high performance across different environments and malware types. This research can contribute to detecting malware attacks and can be integrated in threat monitoring systems. The successful application of this hybrid model indicates its potential for deployment in real-world cybersecurity environments, providing a strong defense against evolving malware threats.

Test Input Prioritization for 3D Point Clouds

3D point cloud applications have become increasingly prevalent in diverse domains, showcasing their efficacy in various software systems. However, testing such applications presents unique challenges due to the high-dimensional nature of 3D point cloud data and the vast number of possible test cases. Test input prioritization has emerged as a promising approach to enhance testing efficiency by prioritizing potentially misclassified test cases during the early stages of the testing process. Consequently, this enables the early labeling of critical inputs, leading to a reduction in the overall labeling cost. However, applying existing prioritization methods to 3D point cloud data is constrained by several factors: (1) inadequate consideration of crucial spatial information, and (2) susceptibility to noises inherent in 3D point cloud data. In this article, we propose PCPrior , the first test prioritization approach specifically designed for 3D point cloud test cases. The fundamental concept behind PCPrior is that test inputs closer to the decision boundary of the model are more likely to be predicted incorrectly. To capture the spatial relationship between a point cloud test and the decision boundary, we propose transforming each test (a point cloud) into a low-dimensional feature vector, toward indirectly revealing the underlying proximity between a test and the decision boundary. To achieve this, we carefully design a group of feature generation strategies, and for each test input, we generate four distinct types of features, namely spatial features, mutation features, prediction features, and uncertainty features. Through a concatenation of the four feature types, PCPrior assembles a final feature vector for each test. Subsequently, a ranking model is employed to estimate the probability of misclassification for each test based on its feature vector. Finally, PCPrior ranks all tests based on their misclassification probabilities. We conducted an extensive study based on 165 subjects to evaluate the performance of PCPrior, encompassing both natural and noisy datasets. The results demonstrate that PCPrior outperforms all of the compared test prioritization approaches, with an average improvement of 10.99% to 66.94% on natural datasets and 16.62% to 53% on noisy datasets.

Biomedical named entity recognition based on multi-cross attention feature fusion.

Currently, in the field of biomedical named entity recognition, CharCNN (Character-level Convolutional Neural Networks) or CharRNN (Character-level Recurrent Neural Network) is typically used independently to extract character features. However, this approach does not consider the complementary capabilities between them and only concatenates word features, ignoring the feature information during the process of word integration. Based on this, this paper proposes a method of multi-cross attention feature fusion. First, DistilBioBERT and CharCNN and CharLSTM are used to perform cross-attention word-char (word features and character features) fusion separately. Then, the two feature vectors obtained from cross-attention fusion are fused again through cross-attention to obtain the final feature vector. Subsequently, a BiLSTM is introduced with a multi-head attention mechanism to enhance the model's ability to focus on key information features and further improve model performance. Finally, the output layer is used to output the final result. Experimental results show that the proposed model achieves the best F1 values of 90.76%, 89.79%, 94.98%, 80.27% and 88.84% on NCBI-Disease, BC5CDR-Disease, BC5CDR-Chem, JNLPBA and BC2GM biomedical datasets respectively. This indicates that our model can capture richer semantic features and improve the ability to recognize entities.

Hypergraph Isomorphism Computation.

The isomorphism problem, crucial in network analysis, involves analyzing both low-order and high-order structural information. Graph isomorphism algorithms focus on structural equivalence to simplify solver space, aiding applications like protein design, chemical pathways, and community detection. However, they fall short in capturing complex high-order relationships, unlike hypergraph isomorphism methods. Traditional hypergraph methods face challenges like high memory use and inaccurate identification, leading to poor performance. To overcome these, we introduce a hypergraph Weisfeiler-Lehman (WL) test algorithm, extending the WL test from graphs to hypergraphs, and develop a hypergraph WL kernel framework with two variants: the Hypergraph WL Subtree Kernel and Hypergraph WL Hyperedge Kernel. The Hypergraph WL Subtree Kernel counts different types of rooted subtrees and generates the final feature vector for a given hypergraph by comparing the number of different types of rooted subtrees. The Subtree Kernel identifies different rooted subtrees, while the Hyperedge Kernel focuses on hyperedges' vertex labels, enhancing feature vector generation. In order to fulfill our research objectives, a comprehensive set of experiments was meticulously designed, including seven graph classification datasets and 12 hypergraph classification datasets. Results on graph classification datasets indicate that the Hypergraph WL Subtree Kernel can achieve the same performance compared with the classical Graph Weisfeiler-Lehman Subtree Kernel. Results on hypergraph classification datasets show significant improvements compared to other typical kernel-based methods, which demonstrates the effectiveness of the proposed methods. In our evaluation, our proposed methods outperform the second-best method in terms of runtime, running over 80 times faster when handling complex hypergraph structures. This significant speed advantage highlights the great potential of our methods in real-world applications.

A Knowledge Graph Completion Algorithm Based on the Fusion of Neighborhood Features and vBiLSTM Encoding for Network Security

Knowledge graphs in the field of network security can integrate diverse, heterogeneous, and fragmented network security data, further explore the relationships between data, and provide support for deep analysis. Currently, there is sparse security information in the field of network security knowledge graphs. The limited information provided by traditional text encoding models leads to insufficient reasoning ability, greatly restricting the development of this field. Starting from text encoding, this paper first addresses the issue of the inadequate capabilities of traditional models using a deep learning model for assistance. It designs a vBiLSTM model based on a word2vec and BiLSTM combination to process network security texts. By utilizing word vector models to retain semantic information in entities and extract key features to input processed data into BiLSTM networks for extracting higher-level features that better capture and express their deeper meanings, this design significantly enhances understanding and expression capabilities toward complex semantics in long sentences before inputting final feature vectors into the KGC-N model. The KGC-N model uses feature vectors combined with graph structure information to fuse forward and reverse domain features and then utilizes a Transformer decoder to decode predictions and complete missing information within the network security knowledge map. Compared with other models using evaluation metrics such as MR, MRR demonstrates that employing our proposed method effectively improves performance on completion tasks and increases comprehension abilities toward complex relations, thereby enhancing accuracy and efficiency when completing knowledge graphs.

Large vessel occlusion detection by non-contrast CT using artificial ıntelligence

IntroductionComputer vision models have been used to diagnose some disorders using computer tomography (CT) and magnetic resonance (MR) images. In this work, our objective is to detect large and small brain vessel occlusion using a deep feature engineering model in acute of ischemic stroke.MethodsWe use our dataset. which contains 324 patient’s CT images with two classes; these classes are large and small brain vessel occlusion. We divided the collected image into horizontal and vertical patches. Then, pretrained AlexNet was utilized to extract deep features. Here, fc6 and fc7 (sixth and seventh fully connected layers) layers have been used to extract deep features from the created patches. The generated features from patches have been concatenated/merged to generate the final feature vector. In order to select the best combination from the generated final feature vector, an iterative selector (iterative neighborhood component analysis—INCA) has been used, and this selector has chosen 43 features. These 43 features have been used for classification. In the last phase, we used a kNN classifier with tenfold cross-validation.ResultsBy using 43 features and a kNN classifier, our AlexNet-based deep feature engineering model surprisingly attained 100% classification accuracy.ConclusionThe obtained perfect classification performance clearly demonstrated that our proposal could separate large and small brain vessel occlusion detection in non-contrast CT images. In this aspect, this model can assist neurology experts with the early recanalization chance.

Deep learning based biometric authentication using electrocardiogram and iris

Authentication systems play an important role in wide range of applications. The traditional token certificate and password-based authentication systems are now replaced by biometric authentication systems. Generally, these authentication systems are based on the data obtained from face, iris, electrocardiogram (ECG), fingerprint and palm print. But these types of models are unimodal authentication, which suffer from accuracy and reliability issues. In this regard, multimodal biometric authentication systems have gained huge attention to develop the robust authentication systems. Moreover, the current development in deep learning schemes have proliferated to develop more robust architecture to overcome the issues of tradition machine learning based authentication systems. In this work, we have adopted ECG and iris data and trained the obtained features with the help of hybrid convolutional neural network- long short-term memory (CNN-LSTM) model. In ECG, R peak detection is considered as an important aspect for feature extraction and morphological features are extracted. Similarly, gabor-wavelet, gray level co-occurrence matrix (GLCM), gray level difference matrix (GLDM) and principal component analysis (PCA) based feature extraction methods are applied on iris data. The final feature vector is obtained from MIT-BIH and IIT Delhi Iris dataset which is trained and tested by using CNN-LSTM. The experimental analysis shows that the proposed approach achieves average accuracy, precision, and F1-core as 0.985, 0.962 and 0.975, respectively.

Breast Tumor Classification with Enhanced Transfer Learning Features and Selection Using Chaotic Map-Based Optimization

Among women, breast cancer remains one of the most dominant cancer types. In the year 2022, around 2,87,800 new cases were diagnosed, and 43,200 women faced mortality due to this disease. Analysis and processing of mammogram images is vital for its earlier identification and thus helps in reducing mortality rates and facilitating effective treatment for women. Accordingly, several deep-learning techniques have emerged for mammogram classification. However, it is still challenging and requires promising solutions. This study proposed a newer automated computer-aided implementation for breast cancer classification. The work starts with enhancing the mammogram contrast using a haze-reduced adaptive technique followed by augmentation. Afterward, EfficientNet-B4 pre-trained architecture is trained for both original and enhanced sets of mammograms individually using static hyperparameters’ initialization. This provides an output of 1792 feature vectors for each set and then fused using a serial mid-value-based approach. The final feature vectors are then optimized using a chaotic-crow-search optimization algorithm. Finally, the obtained significant feature vectors were classified with the aid of machine learning algorithms. The evaluation is made using INbreast and CBIS-DDSM databases. The proposed framework attained a balanced computation time with a maximum classification performance of 98.459 and 96.175% accuracies on INbreast and CBIS-DDSM databases, respectively.

Content-based image retrieval using multi-scale averaging local binary patterns

Feature extraction has a significant impact on the accuracy of Content-Based Image Retrieval (CBIR) algorithms since the content of images is encoded in feature vectors. In this paper, an effective method for texture feature extraction is proposed based on local patterns. In the proposed method, first the image is formed in different scales, then the texture features are extracted from the scales of the image. Finally, extracted features of different scales are concatenated to construct the final feature vector. To evaluate the proposed method, five datasets including Corel-1k, Brodatz, VisTex, Corel-10k, STex, Caltech256, and Oliva are used. In the evaluation process, the effect of different scales and different coding schemes on retrieval precision is investigated. The proposed method is compared with existing CBIR models based on local patterns. The proposed method achieves the best precision of 64.16%, 81.66%, 88.59%, 30.63%, and 69.63%, 13.59%, and 64.10% on the Corel-1k, Brodatz, VisTex, Corel-10k, STex, Caltech256, and Oliva datasets, respectively.

MIFAM-DTI: a drug-target interactions predicting model based on multi-source information fusion and attention mechanism.

Accurate identification of potential drug-target pairs is a crucial step in drug development and drug repositioning, which is characterized by the ability of the drug to bind to and modulate the activity of the target molecule, resulting in the desired therapeutic effect. As machine learning and deep learning technologies advance, an increasing number of models are being engaged for the prediction of drug-target interactions. However, there is still a great challenge to improve the accuracy and efficiency of predicting. In this study, we proposed a deep learning method called Multi-source Information Fusion and Attention Mechanism for Drug-Target Interaction (MIFAM-DTI) to predict drug-target interactions. Firstly, the physicochemical property feature vector and the Molecular ACCess System molecular fingerprint feature vector of a drug were extracted based on its SMILES sequence. The dipeptide composition feature vector and the Evolutionary Scale Modeling -1b feature vector of a target were constructed based on its amino acid sequence information. Secondly, the PCA method was employed to reduce the dimensionality of the four feature vectors, and the adjacency matrices were constructed by calculating the cosine similarity. Thirdly, the two feature vectors of each drug were concatenated and the two adjacency matrices were subjected to a logical OR operation. And then they were fed into a model composed of graph attention network and multi-head self-attention to obtain the final drug feature vectors. With the same method, the final target feature vectors were obtained. Finally, these final feature vectors were concatenated, which served as the input to a fully connected layer, resulting in the prediction output. MIFAM-DTI not only integrated multi-source information to capture the drug and target features more comprehensively, but also utilized the graph attention network and multi-head self-attention to autonomously learn attention weights and more comprehensively capture information in sequence data. Experimental results demonstrated that MIFAM-DTI outperformed state-of-the-art methods in terms of AUC and AUPR. Case study results of coenzymes involved in cellular energy metabolism also demonstrated the effectiveness and practicality of MIFAM-DTI. The source code and experimental data for MIFAM-DTI are available at https://github.com/Search-AB/MIFAM-DTI.

Multi-biometric fusion for enhanced human authentication in information security

In the digital age, information security has taken on greater importance, necessitating the development of strong and trustworthy human authentication techniques. Traditional single-factor authentication schemes, such passwords or PINs, are prone to a number of security flaws. Multi-biometric fusion approaches have become a possible remedy to these restrictions. In comparison to other fusion techniques, feature level fusion is a very popular technique. Features are taken from every biometric attribute in this fusion. Then, the retrieved features are concatenated to create a high dimension final feature vector. In this study, we present a novel method for feature-level fusion employing optimal feature level fusion. The essential features are chosen and identified using a Binary chimp optimized adaptive kernel support vector machine (BCO-AKSVM). According to f1-score (99 %), recall (98 %), accuracy (97 %), precision (88 %), and time computation (1000 ms), the experimental results of the suggested method are analyzed. When compared to current methods, it can show that the proposed BCO-AKSVM achieves the highest performance of information security in human biometric authentication. Multi-biometric fusion is anticipated to play a significant role as technology develops in maintaining secure access control and safeguarding sensitive information in numerous sectors.

Fusion of Attention-Based Convolution Neural Network and HOG Features for Static Sign Language Recognition

The deaf and hearing-impaired community expresses their emotions, communicates with society, and enhances the interaction between humans and computers using sign language gestures. This work presents a strategy for efficient feature extraction that uses a combination of two different methods that are the convolutional block attention module (CBAM)-based convolutional neural network (CNN) and standard handcrafted histogram of oriented gradients (HOG) feature descriptor. The proposed framework aims to enhance accuracy by extracting meaningful features and resolving issues like rotation, similar hand orientation, etc. The HOG feature extraction technique provides a compact feature representation that signifies meaningful information about sign gestures. The CBAM attention module is incorporated into the structure of CNN to enhance feature learning using spatial and channel attention mechanisms. Then, the final feature vector is formed by concatenating these features. This feature vector is provided to the classification layers to predict static sign gestures. The proposed approach is validated on two publicly available static Massey American Sign Language (ASL) and Indian Sign Language (ISL) databases. The model’s performance is evaluated using precision, recall, F1-score, and accuracy. Our proposed methodology achieved 99.22% and 99.79% accuracy for the ASL and ISL datasets. The acquired results signify the efficiency of the feature fusion and attention mechanism. Our network performed better in accuracy compared to the earlier studies.

GCLP: An automated asthma detection model based on global chaotic logistic pattern using cough sounds

Asthmatic patients suffer episodic attacks of breathing difficulty and cough. Cough sounds may be used to distinguish asthma from other lung conditions or normal, especially in the acute period. We aimed to develop a novel architecture using lightweight, handcrafted components that could still extract meaningful features. Cough sounds were recorded using mobile phones from 1428 asthmatic and healthy subjects. As cough sounds behave similarly to random noises, we used a global chaotic logistic pattern (GCLP) as a kernel for feature generation based on its ability to detect random relations among input signal values. Further, each cough sound was preprocessed to remove speech and silent periods before undergoing signal decomposition using tunable Q wavelet transformation (TQWT). The latter enabled downstream GCLP-based feature generation at multiple levels, which yielded four final feature vectors (which corresponded to the four types of TQWT-generated wavelet subbands), each of length 3584. Four feature selectors were applied to the latter, generating 16 selected feature vectors, each of length 256. A standard shallow cubic support vector machine was deployed to calculate 16 prediction vectors, which were collectively input to a mode-based iterative hard majority voting function to generate additional 15-voted results. Finally, a greedy algorithm automatically chose the best overall result, which made the model self-organized. Our model attained 99.44% classification accuracy for the collected dataset. Furthermore, our model obtained an accuracy of 98.53% with leave-one- subject-out cross-validation (LOSO CV) strategy, which justifies the robustness of the developed model.

A Sparse Local Binary Pattern extraction algorithm applied to event sensor data for object classification

Recently, new sensors with active pixels were brought to market. These sensors export local variations of light intensity in the form of asynchronous events with low latency. Since the data output format is a stream of addressable events and not a complete image of light intensities, new algorithms are required for known problems in the field of Computer Vision, such as segmentation, VO, SLAM, object, and scene recognition. There are some proposed methodologies for object recognition using conventional methods, convolutional neural networks, and third-generation neural networks based on spikes. However, convolutional neural networks and spike neural networks require specific hardware for processing, hard to miniaturize. Also, several traditional Computer Vision operators and feature descriptors have been neglected in the context of event sensors and could contribute to lighter methodologies in object recognition. This paper proposes an algorithm for local binary pattern extraction in sparse structures, typically found in this context. This paper also proposes two methodologies using local binary patterns to captures with event-based sensors for object recognition. The first methodology exploits the known motion performed by the sensor, while the second is motion agnostic. It is demonstrated experimentally that the LBP operator is a fast and light alternative that enables variable reduction using PCA in some cases. The experiments also show that it is possible to reduce the final feature vector for classification by up to 99,73% when compared to conventional methods considered state-of-the-art while maintaining comparable accuracy.

Tell me your position: Distantly supervised biomedical entity relation extraction using entity position marker

A significant amount of textual data has been produced in the biomedical area recently as a result of the advancement of biomedical technologies. Large-scale biomedical data can be automatically obtained with the help of distant supervision. However, the noisy data brought by distant supervision methods makes relation extraction tasks more difficult. Previous work has focused more on how to restore mislabeled relationships, but little attention has been paid to the importance of labeled entity locations for relationship extraction tasks. In this paper, we present a “four-stage” model based on BioBERT and Multi-Instance Learning by using entity position markers. Firstly, the sentence is marked with position. Secondly, BioBERT, a biomedical pre-trained language model, is used in the final sentence feature vector representation not only with the global position marker but also with the start and end marker of both the head and tail entity. Thirdly, the aggregation of sentence vectors in the bag is used as the vector feature of the bag by three aggregation methods, and the performance of different sentence feature vectors combined with different bag encoding methods is discussed. At last, relation classification is performed at the bag level. According to experimental results, the presented model significantly outperforms all baseline models and contributes to noise reduction. In addition, different bag encoding methods need to match corresponding sentence encoding representation to achieve the best performance.

NFSDense201: microstructure image classification based on non-fixed size patch division with pre-trained DenseNet201 layers

In the field of nanoscience, the scanning electron microscope (SEM) is widely employed to visualize the surface topography and composition of materials. In this study, we present a novel SEM image classification model called NFSDense201, which incorporates several key components. Firstly, we propose a unique nested patch division approach that divides each input image into four patches of varying dimensions. Secondly, we utilize DenseNet201, a deep neural network pretrained on ImageNet1k, to extract 2920 deep features from the last fully connected and global average pooling layers. Thirdly, we introduce an iterative neighborhood component analysis function to select the most discriminative features from the merged feature vector, which is formed by concatenating the four feature vectors extracted per input image. This process results in a final feature vector of optimal length 698. Lastly, we employ a standard shallow support vector machine classifier to perform the actual classification. To evaluate the performance of NFSDense201, we conducted experiments using a large public SEM image dataset. The dataset consists of 972, 162, 326, 4590, 3820, 3925, 4755, 181, 917, and 1624.jpeg images belonging to the following microstructural categories: “biological,” “fibers,” “film-coated surfaces,” “MEMS devices and electrodes,” “nanowires,” “particles,” “pattern surfaces,” “porous sponge,” “powder,” and “tips,” respectively. For both four-class and ten-class classification tasks, we evaluated NFSDense201 using subsets of the dataset containing 5080 and 21,272 images, respectively. The results demonstrate the superior performance of NFSDense201, achieving a four-class classification accuracy rate of 99.53% and a ten-class classification accuracy rate of 97.09%. These accuracy rates compare favorably against previously published SEM image classification models. Additionally, we report the performance of NFSDense201 for each class in the dataset.

Feature extraction and fusion using deep convolutional neural networks for Covid-19 detection using CT and X-RAY images

COVID-19 represents a novel variant of the coronavirus disease, having rapidly disseminated across the globe. In recent studies pertaining to computer vision, image processing, and classification techniques, numerous methodologies have been introduced employing chest X-ray images and computerized tomography (CT) images. This investigation introduces novel fused deep features founded on CT and X-ray image classification, with the aim of detecting COVID-19 disease. The proposed approach encompasses three key phases: deep feature generation, iterative feature selection, and classification. During the feature generation phase, well-known pre-trained deep convolutional neural networks, namely DenseNet201, MobileNetV2, ResNet18, ResNet50, ResNet101, VGG16, and VGG19, were leveraged. Each network model generated 1000 features, which were subsequently fused, culminating in the acquisition of a final feature vector comprising 7000 dimensions. In order to distill the most pertinent information, the ReliefF and iterative maximum relevance minimum redundancy (RFImRMR) feature selection techniques were employed in the generation of the ultimate feature vector. To evaluate the performance of the proposed method, publicly available datasets of CT images and X-ray images were employed. Notably, the suggested deep learning approach attained accuracy rates of 99.33% and 93.10% for COVID-19 detection using CT and X-ray images, respectively. These achieved results serve as compelling evidence substantiating the efficacy of the proposed fused deep features and RFImRMR-based COVID-19 detection.

"Where does it hurt?": Exploring EDA Signals to Detect and Localise Acute Pain.

Pain is a highly unpleasant sensory experience, for which currently no objective diagnostic test exists to measure it. Identification and localisation of pain, where the subject is unable to communicate, is a key step in enhancing therapeutic outcomes. Numerous studies have been conducted to categorise pain, but no reliable conclusion has been achieved. This is the first study that aims to show a strict relation between Electrodermal Activity (EDA) signal features and the presence of pain and to clarify the relation of classified signals to the location of the pain. For that purpose, EDA signals were recorded from 28 healthy subjects by inducing electrical pain at two anatomical locations (hand and forearm) of each subject. The EDA data were preprocessed with a Discrete Wavelet Transform to remove any irrelevant information. Chi-square feature selection was used to select features extracted from three domains: time, frequency, and cepstrum. The final feature vector was fed to a pool of classification schemes where an Artificial Neural Network classifier performed best. The proposed method, evaluated through leave-one-subject-out cross-validation, provided 90% accuracy in pain detection (no pain vs. pain), whereas the pain localisation experiment (hand pain vs. forearm pain) achieved 66.67% accuracy.Clinical relevance- This is the first study to provide an analysis of EDA signals in finding the source of the pain. This research explores the viability of using EDA for pain localisation, which may be helpful in the treatment of noncommunicable patients.

A DCT probability histogram-based ROI features for content-based natural and medical image retrieval applications

The effectiveness of a Content-based Image Retrieval (CBIR) system depends on selecting representative and salient visual features. Capturing the salient image features for forming a feature vector is an essential part of the CBIR process, as it can significantly affect retrieval efficiency. Most images contain a visually prominent region that perceives maximum image semantic information, known as Regions-of-Interest (ROIs). This paper suggests a new CBIR scheme that employs a Graph-based Visual Saliency (GBVS) map to extract the ROI of the image. Further, block-level Discrete Cosine Transforms (DCT) and subsequent histograms are computed for the DC and significant AC coefficients of the extracted ROI image. In addition, a low-dimensional feature vector is formed from the modified histograms. Subsequently, six statistical features from the background scene have also been computed for the final feature vector construction. As a result, the final feature vector contains ROI and background information in their proper proportion to improve retrieval performance. Moreover, the performance of the proposed CBIR scheme has been evaluated using one medical, three natural, two objects, one texture, and one natural heterogeneous image database. All these databases collectively contain 53732 different images from 200 image classes. The outcomes of the experiments demonstrate substantial improvement in the retrieval process with some existing related schemes. At the same time, it also shows that the performance of the proposed scheme is robust concerning all kinds of image databases.

Sentiment Analysis of the Consumer Review Text Based on BERT-BiLSTM in a Social Media Environment

In this paper, a BERT-BiLSTM-based consumer review text sentiment analysis method in the e-commerce big data field is proposed. First, the unlabeled text is trained using the BERT training model for the language introduced in the deep learning, and then the pre-training model of the text data is delivered by the learning textual features and data to extract deeper vectors. Second, the BiLSTM model is applied to simultaneously obtain contextual information so as to illustrate optimal textual features. Finally, a corresponding sentiment analysis model relative to the consumer review text is constructed by combining the BERT model with BiLSTM to better merge the context for classifying sentiment and improving the final feature vector accuracy for the sentiment classification results. Simulated by experiments, the method proposed in this paper was compared with another three methods using the same data set. The results obtained indicate that the proposed method has the highest precision, recall, and F1-Measure, and the values reach 92.64%, 90.32%, and 91.46%, respectively.