Traditional Deep Learning Models Research Articles

Detection of peach soluble solids based on near-infrared spectroscopy with High Order Spatial Interaction network.

Due to the scalability of deep learning technology, researchers have applied it to the non-destructive testing of peach internal quality. In addition, the soluble solids content (SSC) is an important internal quality indicator that determines the quality of peaches. Peaches with high SSC have a sweeter taste and better texture, making them popular in the market. Therefore, SSC is an important indicator for measuring peach internal quality and making harvesting decisions. This article presents the High Order Spatial Interaction Network (HOSINet), which combines the Position Attention Module (PAM) and Channel Attention Module (CAM). Additionally, a feature wavelength selection algorithm similar to the Group-based Clustering Subspace Representation (GCSR-C) is used to establish the Position and Channel Attention Module-High Order Spatial Interaction (PC-HOSI) model for peach SSC prediction. The accuracy of this model is compared with traditional machine learning and traditional deep learning models. Finally, the permutation algorithm is combined with deep learning models to visually evaluate the importance of feature wavelengths. Increasing the order of the PC-HOSI model enhances its ability to learn spatial correlations in the dataset, thus improving its predictive performance. The optimal model, PC-HOSI model, performed well with an order of 3 (PC-HOSI-3), with a root mean square error of 0.421 °Brix and a coefficient of determination of 0.864. Compared with traditional machine learning and deep learning algorithms, the coefficient of determination for the prediction set was improved by 0.07 and 0.39, respectively. The permutation algorithm also provided interpretability analysis for the predictions of the deep learning model, offering insights into the importance of spectral bands. These results contribute to the accurate prediction of SSC in peaches and support research on interpretability of neural network models for prediction. © 2024 Society of Chemical Industry.

DiffusionFR: Species Recognition of Fish in Blurry Scenarios via Diffusion and Attention

Blurry scenarios, such as light reflections and water ripples, often affect the clarity and signal-to-noise ratio of fish images, posing significant challenges for traditional deep learning models in accurately recognizing fish species. Firstly, deep learning models rely on a large amount of labeled data. However, it is often difficult to label data in blurry scenarios. Secondly, existing deep learning models need to be more effective for the processing of bad, blurry, and otherwise inadequate images, which is an essential reason for their low recognition rate. A method based on the diffusion model and attention mechanism for fish image recognition in blurry scenarios, DiffusionFR, is proposed to solve these problems and improve the performance of species recognition of fish images in blurry scenarios. This paper presents the selection and application of this correcting technique. In the method, DiffusionFR, a two-stage diffusion network model, TSD, is designed to deblur bad, blurry, and otherwise inadequate fish scene pictures to restore clarity, and a learnable attention module, LAM, is intended to improve the accuracy of fish recognition. In addition, a new dataset of fish images in blurry scenarios, BlurryFish, was constructed and used to validate the effectiveness of DiffusionFR, combining bad, blurry, and otherwise inadequate images from the publicly available dataset Fish4Knowledge. The experimental results demonstrate that DiffusionFR achieves outstanding performance on various datasets. On the original dataset, DiffusionFR achieved the highest training accuracy of 97.55%, as well as a Top-1 accuracy test score of 92.02% and a Top-5 accuracy test score of 95.17%. Furthermore, on nine datasets with light reflection noise, the mean values of training accuracy reached a peak at 96.50%, while the mean values of the Top-1 accuracy test and Top-5 accuracy test were at their highest at 90.96% and 94.12%, respectively. Similarly, on three datasets with water ripple noise, the mean values of training accuracy reached a peak at 95.00%, while the mean values of the Top-1 accuracy test and Top-5 accuracy test were at their highest at 89.54% and 92.73%, respectively. These results demonstrate that the method showcases superior accuracy and enhanced robustness in handling original datasets and datasets with light reflection and water ripple noise.

Autonomous Driving Control for Passing Unsignalized Intersections Using the Semantic Segmentation Technique

Autonomous driving in urban areas is challenging because it requires understanding vehicle movements, traffic rules, map topologies and unknown environments in the highly complex driving environment, and thus typical urban traffic scenarios include various potentially hazardous situations. Therefore, training self-driving cars by using traditional deep learning models not only requires the labelling of numerous datasets but also takes a large amount of time. Because of this, it is important to find better alternatives for effectively training self-driving cars to handle vehicle behavior and complex road shapes in dynamic environments and to follow line guidance information. In this paper, we propose a method for training a self-driving car in simulated urban traffic scenarios to be able to judge the road conditions on its own for crossing an unsignalized intersection. In order to identify the behavior of traffic flow at the intersection, we use the CARLA (CAR Learning to Act) self-driving car simulator to build the intersection environment and simulate the process of traffic operation. Moreover, we attempt to use the DDPG (Deep Deterministic Policy Gradient) and RDPG (Recurrent Deterministic Policy Gradient) learning algorithms of the DRL (Deep Reinforcement Learning) technology to train models based on the CNN (Convolutional Neural Network) architecture. Specifically, the observation image of the semantic segmentation camera installed on the self-driving car and the vehicle speed are used as the model input. Moreover, we design an appropriate reward mechanism for performing training according to the current situation of the self-driving car judged from sensing data of the obstacle sensor, collision sensor and lane invasion detector. Doing so can improve the convergence speed of the model to achieve the purpose of the self-driving car autonomously judging the driving paths so as to accomplish accurate and stable autonomous driving control.

A robust intrusion detection system based on a shallow learning model and feature extraction techniques.

The escalating prevalence of cybersecurity risks calls for a focused strategy in order to attain efficient resolutions. This study introduces a detection model that employs a tailored methodology integrating feature selection using SHAP values, a shallow learning algorithm called PV-DM, and machine learning classifiers like XGBOOST. The efficacy of our suggested methodology is highlighted by employing the NSL-KDD and UNSW-NB15 datasets. Our approach in the NSL-KDD dataset exhibits exceptional performance, with an accuracy of 98.92%, precision of 98.92%, recall of 95.44%, and an F1-score of 96.77%. Notably, this performance is achieved by utilizing only four characteristics, indicating the efficiency of our approach. The proposed methodology achieves an accuracy of 82.86%, precision of 84.07%, recall of 77.70%, and an F1-score of 80.20% in the UNSW-NB15 dataset, using only six features. Our research findings provide substantial evidence of the enhanced performance of the proposed model compared to a traditional deep-learning model across all performance metrics.

Ensemble Learning Method for the Continuous Decoding of Hand Joint Angles.

Human-machine interface technology is fundamentally constrained by the dexterity of motion decoding. Simultaneous and proportional control can greatly improve the flexibility and dexterity of smart prostheses. In this research, a new model using ensemble learning to solve the angle decoding problem is proposed. Ultimately, seven models for angle decoding from surface electromyography (sEMG) signals are designed. The kinematics of five angles of the metacarpophalangeal (MCP) joints are estimated using the sEMG recorded during functional tasks. The estimation performance was evaluated through the Pearson correlation coefficient (CC). In this research, the comprehensive model, which combines CatBoost and LightGBM, is the best model for this task, whose average CC value and RMSE are 0.897 and 7.09. The mean of the CC and the mean of the RMSE for all the test scenarios of the subjects' dataset outperform the results of the Gaussian process model, with significant differences. Moreover, the research proposed a whole pipeline that uses ensemble learning to build a high-performance angle decoding system for the hand motion recognition task. Researchers or engineers in this field can quickly find the most suitable ensemble learning model for angle decoding through this process, with fewer parameters and fewer training data requirements than traditional deep learning models. In conclusion, the proposed ensemble learning approach has the potential for simultaneous and proportional control (SPC) of future hand prostheses.

Short Text Semantic Sentiment Analysis Based on Dual Channel Aspect Attention in Intelligent Systems

Traditional deep learning models for text sentiment analysis fail to fully harness the contextual semantic information of aspect nodes or use prior sentiment resources. This paper proposes a dual channel sentiment analysis model named M2BERT-BLSTM AA that is based on an enhanced Bidirectional Encoder Representations from Transformers(BERT)and Bidirectional Long short-term memory(BLSTM) model and incorporates a Dual Attention Mechanism. Firstly, an emotional resource database is constructed using existing emotional resources. Secondly, vectors are concatenated following mean and max pooling along the dimension of sentence length. These semantic features mitigate evaluation imbalance.Then the text and sentiment information are encoded separately, using distinct Attention Mechanism(Att-M) to extract contextual relationships and emotional features. The model's Aspect-Based multi-class sentiment prediction accuracies on the three Chinese datasets of Meituan ordering, restaurants, and laptops are 75.2%, 87.5%, and 75%, respectively, showing improved performance on sentiment classification.

Forecast of solar cycle 25 based on Hybrid CNN-Bidirectional-GRU (CNN-BiGRU) model and Novel Gradient Residual Correction (GRC) technique

In this study, a hybrid deep-learning(DL) model is proposed, which consists of a Convolution-Neural-Network (CNN) and Bidirectional-Gated-Recurrent-Unit (Bi-GRU) for the prediction of sunspot numbers(SSN) of different frequencies along with novel post-processing techniques of Gradient-Residual-Correction (GRC) to enhance the accuracy of the predictions further. GRC is utilized to reduce the residual present in the forecasted values. The AdaBoost regression model is implemented over the residual obtained from the prediction of training data using a hybrid CNN-BiGRU model with respect to the gradient of the training data points to predict the residual for the test data points. Ultimately, the predicted residual for test data points is summed up with the predicted test data points to achieve the final predictions. The results obtained from the proposed methods are compared with the results obtained from traditional DL models. The validation of the proposed method is carried out based on four performance metrics, namely Root Mean Squared Error(RMSE), Mean Absolute Scaled Error(MASE), Mean Absolute Error(MAE), and Mean Absolute Percentage Error(MAPE). A significant percentage of improvement is observed while using the GRC technique in comparison to all other experimented models for all four variants of SSN data. A maximum improvement of 85.71% has been achieved in comparison to the BiLSTM model over the 13-month smoothed SSN dataset on the basis of MAPE. Friedman Ranking is also performed as a non-parametric statistical test over the results of the performance measures. This model has been utilized for the forecast of solar cycle 25(SC25) over the annual mean of total SSN. It has been observed that the SC25 is expected to reach its peak in the year 2024 with an annual average peak value of 143.641. Comparative analysis of SC25 and the peak of SSN in the ongoing cycle with the previous works are also carried out.

Continual Learning with Deep Neural Networks in Physiological Signal Data: A Survey.

Deep-learning algorithms hold promise in processing physiological signal data, including electrocardiograms (ECGs) and electroencephalograms (EEGs). However, healthcare often requires long-term monitoring, posing a challenge to traditional deep-learning models. These models are generally trained once and then deployed, which limits their ability to adapt to the dynamic and evolving nature of healthcare scenarios. Continual learning-known for its adaptive learning capabilities over time-offers a promising solution to these challenges. However, there remains an absence of consolidated literature, which reviews the techniques, applications, and challenges of continual learning specific to physiological signal analysis, as well as its future directions. Bridging this gap, our review seeks to provide an overview of the prevailing techniques and their implications for smart healthcare. We delineate the evolution from traditional approaches to the paradigms of continual learning. We aim to offer insights into the challenges faced and outline potential paths forward. Our discussion emphasizes the need for benchmarks, adaptability, computational efficiency, and user-centric design in the development of future healthcare systems.

CT-Based Super-Resolution Deep Learning Models with Attention Mechanisms for Predicting Spread Through Air Spaces of Solid or Part-Solid Lung Adenocarcinoma

Rationale and ObjectivesSpread through air space (STAS) is a novel invasive pattern of lung adenocarcinoma (LUAD), and preoperative knowledge of STAS status is helpful in choosing an appropriate surgical approach. Materials and MethodsThis retrospective study collected and analyzed 602 patients diagnosed with LUAD from two medical centers: Center 1 was randomly partitioned into training (n = 358) and validation cohorts (n = 154) at a 7:3 ratio; and center 2 was the external test cohort (n = 90). Super resolution was performed on all images to acquire high-resolution images, which were used to train the SE-ResNet50 model, before creating an equivalent parameter ResNet50 model. Disparities were compared between the two models using receiver operating characteristic curves, area under the curve, accuracy, precision, sensitivity, and specificity. ResultsIn this study, 512 and 90 patients with LUAD were enrolled from centers 1 and 2, respectively. The curve values of the SE-ResNet50 and ResNet50 models were compared for training, validation, and test cohorts, resulting in values of 0.933 vs. 0.909, 0.783 vs. 0.728, and 0.806 vs. 0.695, respectively. In the external test cohort, the accuracy of the SE-ResNet50 model demonstrated a 10% improvement over the ResNet50 model (82.2% vs. 72.2%). ConclusionThe SE-ResNet50 model based on computed tomography super-resolution has great potential for predicting STAS status in patients with solid or partially solid LUAD, with superior predictive performance compared to traditional deep learning models.

Finsformer: A Novel Approach to Detecting Financial Attacks Using Transformer and Cluster-Attention

This paper aims to address the increasingly severe security threats in financial systems by proposing a novel financial attack detection model, Finsformer. This model integrates the advanced Transformer architecture with the innovative cluster-attention mechanism, dedicated to enhancing the accuracy of financial attack behavior detection to counter complex and varied attack strategies. A key innovation of the Finsformer model lies in its effective capture of key information and patterns within financial transaction data. Comparative experiments with traditional deep learning models such as RNN, LSTM, Transformer, and BERT have demonstrated that Finsformer excels in key metrics such as precision, recall, and accuracy, achieving scores of 0.97, 0.94, and 0.95, respectively. Moreover, ablation studies on different feature extractors further confirm the effectiveness of the Transformer feature extractor in processing complex financial data. Additionally, it was found that the model’s performance heavily depends on the quality and scale of data and may face challenges in computational resources and efficiency in practical applications. Future research will focus on optimizing the Finsformer model, including enhancing computational efficiency, expanding application scenarios, and exploring its application on larger and more diversified datasets.

SBiLM: Siamese Bi-LSTM model for handling imbalance in fake review detection

The growing significance of online reviews in e-commerce platforms has a profound impact on consumers' purchase decisions. Unfortunately, dishonest sellers exploit this system by engaging in deceptive practices and manipulating reviews. These fake reviews are very hard to be detected by humans. Hence, the issue of detecting fake online reviews has gained significant attention due to their prevalence. Typically, the datasets available have much higher number of real reviews as compared to fake reviews, which leads to data imbalance problem. Machine learning algorithms struggle to learn from these datasets and give biased results. However, existing studies on this topic have not adequately addressed the challenges posed by imbalanced data. They have mostly used under-sampling or over-sampling to handle imbalance in datasets, which leads to potential information loss or overfitting due to replication. Hence, to handle this class imbalance problem without using sampling techniques, this paper proposes a novel SBiLM model using Siamese Neural Network with Bidirectional LSTM and Manhattan Distance. SBiLM takes pair of reviews as inputs and computes similarity between them using distance metric. The outcomes indicate that the SBiLM model exhibits better performance in terms of both accuracy and loss, when compared with traditional deep learning models on a highly imbalanced restaurant review dataset. This presents a promising strategy for tackling the issues related to data imbalance in the domain of fake review detection.

An interpretable horizontal federated deep learning approach to improve short-term solar irradiance forecasting

Solar irradiance forecasting is critical in the planning and operation of solar power plants for production scheduling, energy trading, and maintenance planning. Accurate solar irradiance forecasting is crucial for efficient and stable power systems. Various deep learning (DL) methods have been researched in solar irradiance forecasting because of their powerful capabilities of nonlinear mapping. However, traditional DL models face challenges with small sample sizes, irradiance fluctuations, and data privacy concerns. On this foundation, it is also critical to ensure the interpretability of artificial intelligence (AI) models, as interpretable and trustworthy predictive models can help operators make effective decisions. Focusing on the above-mentioned issues, a privacy-preserving, interpretable, and DL-based model is proposed to improve the forecast accuracy of solar irradiance while ensuring the data available but not visible to each participant and prediction comprehensibility. This model is constructed by horizontal federated framework (HFF), self-attention (SA) mechanism, convolutional neural network (CNN), and long short-term memory (LSTM) network, named HFF-based SA-CNN-LSTM. A series of comprehensive experiments spanning various climate regions are designed to assess model performance under different federated conditions. The experiments show that this method significantly improves forecasting accuracy, especially in scenarios with incomplete data and highly fluctuating irradiance when federated with other datasets, and the model exhibits interpretability, adjusting attention for precise predictions at different time points. Moreover, the effects of different federated scales on the predictive performance of the target model are fully discussed for future practical applications. These findings hold significant promise for enhancing the accuracy of irradiance forecasting, contributing to energy efficiency, grid stability, and data privacy.

Improved detection of aortic dissection in non-contrast-enhanced chest CT using an attention-based deep learning model

Rationale and objectivesThis study investigated the effects of implementing an attention-based deep learning model for the detection of aortic dissection (AD) using non-contrast-enhanced chest computed tomography (CT). Materials and methodsWe analysed the records of 1300 patients who underwent contrast-enhanced chest CT at 2 medical centres between January 2015 and February 2023. We considered an internal cohort of 200 patients with AD and 200 patients without AD and an external test cohort of 40 patients with AD and 40 patients without AD. The internal cohort was divided into training and test sets, and a deep learning model was trained using 9600 CT images. A convolutional block attention module (CBAM) and a traditional deep learning architecture (namely, You Only Look Once version 5 [YOLOv5]) were combined into an attention-based model (i.e., YOLOv5-CBAM). Its performance was measured against the unmodified YOLOv5 model, and the accuracy, sensitivity, and specificity of the algorithm were evaluated by two independent radiologists. ResultsThe CBAM-based model outperformed the traditional deep learning model. In the external testing set, YOLOv5-CBAM achieved an area under the curve (AUC) of 0.938, accuracy of 91.5 %, sensitivity of 90.0 %, and specificity of 92.9 %, whereas the unmodified model achieved an AUC of 0.844, accuracy of 83.6 %, sensitivity of 71.2 %, and specificity of 96.0 %. The sensitivity results of the unmodified algorithms were not significantly different from those of the radiologists; however, the proposed YOLOv5-CBAM algorithm outperformed the unmodified algorithms in terms of detection. ConclusionsIncorporating the CBAM attention mechanism into a deep learning model can significantly improve AD detection in non-contrast-enhanced chest CT. This approach may aid radiologists in the timely and accurate diagnosis of AD, which is important for improving patient outcomes.

Experimental evaluation of artificial intelligence assisted heart disease prediction using deep learning principle

The system for detecting cardiac illness utilizing Artificial Intelligence (AI) and deep learning algorithms is the main topic of the study. Researchers demonstrate how artificial intelligence can be used to forecast if someone would get cardiac disease. The goal of this work is to create an artificial intelligence system that can detect cardiac problems using machine learning. Diagnose, risk stratification, and management are essentially some of the essential thinking-intensive elements of healthcare that have been automated due to the creation of Artificial Intelligence along with information science, reducing the burden on doctors and lowering the risk of human error. Clinical decision-support platforms frequently employ artificial intelligence approaches for accurate disease prediction and diagnosis. Considering the health history of the individual, we developed a system to determine if a heart disease diagnosis is likely or not for the patient. In order to demonstrate the effectiveness of the suggested strategy, this research established a revolutionary deep learning-based cardiovascular disease diagnosis logic called Efficient Learning based Health Evaluator (ELHE). It is cross-validated with the traditional deep learning model known as Artificial Neural Network (ANN). Many people with cardiovascular disease have the same blatant warning signs that may be used to make a diagnosis. A scheme of detection built around these risk indicators would benefit healthcare providers as well as patients by alerting them to the potential for heart disease before they enter a hospital or undergo pricey diagnostic tests. Finally, the popular learning-oriented tool Python is used to create the recommended prediction logic known as ELHE, which allows the user to input clinical information and understand the present state of a patient's health. This strategy improves healthcare while lowering the cost of treatment. For the medical professional, this technology will serve as a promising tool for accurate diagnosis.

MC-MLDCNN: Multichannel Multilayer Dilated Convolutional Neural Networks for Web Attack Detection

The explosive growth of web-based technology has led to an increase in sophisticated and complex attacks that target web applications. To protect against this growing threat, a reliable web attack detection methodology is essential. This research aims to provide a method that can detect web attacks accurately. A character-level multichannel multilayer dilated convolutional neural network (MC-MLDCNN) is proposed to identify web attacks accurately. The model receives the full text of HTTP requests as inputs. Character-level embedding is applied to embed HTTP requests to the model. Therefore, feature extraction is carried out automatically by the model, and no additional effort is required. This approach significantly simplifies the preprocessing phase. The methodology consists of multichannel dilated convolutional neural network blocks with various kernel sizes. Each channel involves several layers with exponentially increasing dilation sizes. Through the integration of multichannel and multilayer dilated convolutional neural networks, the model can efficiently capture the temporal relation and dependence of character granularity of HTTP requests at different scales and levels. As a result, the structure enables the model to easily capture dependencies over extended and long sequences of HTTP requests and consequently identify attacks accurately. The outcomes of the experiments carried out on the CSIC 2010 dataset show that the proposed model outperforms several state-of-the-art deep learning-based models in the literature and some traditional deep learning models by identifying web attacks with a precision score of 99.65%, a recall score of 98.80%, an F1 score of 99.22%, and an accuracy score of 99.36%. A useful web attack detection system must be able to balance accurate attack identification with minimizing false positives (identifying normal requests as attacks). The success of the model in recognizing normal requests is further evaluated to guarantee increased security without sacrificing web applications’ usability and availability.

Reconstruction of Continuous High-Resolution Sea Surface Temperature Data Using Time-Aware Implicit Neural Representation

Accurate climate data at fine spatial resolution are essential for scientific research and the development and planning of crucial social systems, such as energy and agriculture. Among them, sea surface temperature plays a critical role as the associated El Niño–Southern Oscillation (ENSO) is considered a significant signal of the global interannual climate system. In this paper, we propose an implicit neural representation-based interpolation method with temporal information (T_INRI) to reconstruct climate data of high spatial resolution, with sea surface temperature as the research object. Traditional deep learning models for generating high-resolution climate data are only applicable to fixed-resolution enhancement scales. In contrast, the proposed T_INRI method is not limited to the enhancement scale provided during the training process and its results indicate that it can enhance low-resolution input by arbitrary scale. Additionally, we discuss the impact of temporal information on the generation of high-resolution climate data, specifically, the influence of the month from which the low-resolution sea surface temperature data are obtained. Our experimental results indicate that T_INRI is advantageous over traditional interpolation methods under different enhancement scales, and the temporal information can improve T_INRI performance for a different calendar month. We also examined the potential capability of T_INRI in recovering missing grid value. These results demonstrate that the proposed T_INRI is a promising method for generating high-resolution climate data and has significant implications for climate research and related applications.

Mortality prediction with adaptive feature importance recalibration for peritoneal dialysis patients

The study aims to develop AICare, an interpretable mortality prediction model, using electronic medical records (EMR) from follow-up visits for end-stage renal disease (ESRD) patients. AICare includes a multichannel feature extraction module and an adaptive feature importance recalibration module. It integrates dynamic records and static features to perform personalized health context representation learning. The dataset encompasses 13,091 visits and demographic data of 656 peritoneal dialysis (PD) patients spanning 12 years. An additional public dataset of 4,789 visits from 1,363 hemodialysis (HD) patients is also considered. AICare outperforms traditional deep learning models in mortality prediction while retaining interpretability. It uncovers mortality-feature relationships and variations in feature importance and provides reference values. An AI-doctor interaction system is developed for visualizing patients' health trajectories and risk indicators.

An adaptive physics-informed deep learning method for pore pressure prediction using seismic data

Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering. Recent deep learning technique has been receiving more interest due to the great potential to deal with pore pressure prediction. However, most of the traditional deep learning models are less efficient to address generalization problems. To fill this technical gap, in this work, we developed a new adaptive physics-informed deep learning model with high generalization capability to predict pore pressure values directly from seismic data. Specifically, the new model, named CGP-NN, consists of a novel parametric features extraction approach (1DCPP), a stacked multilayer gated recurrent model (multilayer GRU), and an adaptive physics-informed loss function. Through machine training, the developed model can automatically select the optimal physical model to constrain the results for each pore pressure prediction. The CGP-NN model has the best generalization when the physics-related metric λ=0.5. A hybrid approach combining Eaton and Bowers methods is also proposed to build machine-learnable labels for solving the problem of few labels. To validate the developed model and methodology, a case study on a complex reservoir in Tarim Basin was further performed to demonstrate the high accuracy on the pore pressure prediction of new wells along with the strong generalization ability. The adaptive physics-informed deep learning approach presented here has potential application in the prediction of pore pressures coupled with multiple genesis mechanisms using seismic data.

Imbalanced sentiment classification of online reviews based on SimBERT

The purpose of sentiment classification is to accomplish automatic judssssgment of the sentiment tendency of text. In the sentiment classification task of online reviews, traditional models focus on the optimization of algorithm performance, but ignore the imbalanced distribution of the number of sentiment classifications of online reviews, which causes serious degradation in the classification performance of the model in practical applications. The experiment was divided into two stages in the overall context. The first stage trains SimBERT using online review data so that SimBERT can fully learn the semantic features of online reviews. The second stage uses the trained SimBERT model to generate fake minority samples and mix them with the original samples to obtain a distributed balanced dataset. Then the mixed data set is input into the deep learning model to complete the sentiment classification task. Experimental results show that this method has excellent classification performance in the sentiment classification task of hotel online reviews compared with traditional deep learning models and models based on other imbalanced processing methods.

Classification of multi-type bearing fault features based on semi-supervised generative adversarial network (GAN)

Fault diagnosis is a crucial technology for ensuring the reliable and efficient operation of industrial systems. With the advancement of industrial informatization and intelligence, fault diagnosis methods have the trend shifting from traditional signal processing to deep learning. However, traditional deep learning models are not suitable for industrial scenarios with limited labeled data, imbalanced categories. To address these challenges, this paper proposes a novel approach based on semi-supervised generative adversarial networks (SGANs) to systematically study the fault diagnosis of rolling bearings in the case of unlabeled samples and sparsely labeled samples. In this work, the vibration time-domain vibration signal of the bearing is firstly transformed into a spectrum signal through the fast Fourier transform. This transformed signal is then fed into the GAN model to extract multi-layer sensitive features, providing a deeper understanding of the underlying fault characteristics. Subsequently, the SGAN method utilizes unsupervised learning via spectral clustering algorithms to automatically classify fault patterns in industrial equipment. Furthermore, it enhances semi-supervised learning by incorporating limited label information through softmax functions, effectively discerning the authenticity of unlabeled data. For the effectiveness of SGAN for bearing fault diagnosis, two diverse datasets are utilized including the widely-used Case Western Reserve University dataset and data acquired from South Ural State University. Compared to alternative models, the results underscore SGAN’s robustness, achieving high recognition accuracy and clustering performance. The proposed methodology contributes to the advancement of fault diagnosis technologies by combining unsupervised and semi-supervised learning techniques.