Applied Soft Computing最新文献

Acquiring labeled data for learning sentence embeddings in Natural Language Processing poses challenges due to limited availability and high costs. In order to tackle this issue, we introduce a novel method called Contrastive Learning with Transformer Initialization and Clustering Prior for Text Representation (CLTC). Our method utilizes Pre-Layernorm Transformers without warm-up, stabilizing the training process while also increasing the final performance. We employ Contrastive Learning (CL) with dropout-based augmentation to enhance sentence embeddings. Additionally, we integrate prior knowledge into the contrastive learning framework within an efficient clustering strategy. When evaluated on the SentEval task, our approach showcases a competitive performance when compared to state-of-the-art approaches in the contrastive learning domain. Our method offers stability, improved embeddings, and the utilization of prior knowledge for enhanced unsupervised representation learning in Natural Language Processing applications.

Coronavirus is an endangered disease to kills more than millions of people, but it has also put tremendous pressure on the whole medical system. The initial stage of identification of COVID-19 is necessary to isolate the patients with positive cases in order to stop the disease from spreading. The amalgamation of imaging techniques and deep learning algorithms takes less time and leads to more accurate outcomes for COVID-19 detection. Deep learning techniques have been employed by scientists to identify coronavirus infection in lung images during the COVID-19 worldwide epidemic. In this review, a review of the Covid-19 detection framework based on machine learning and deep learning techniques using X-ray images is done. First, the review of existing Covid-19 detection models is done. For this purpose, a detailed literature survey is carried out on Covid-19 detection papers from 2019 to 2023. Following the literature survey, the pre-processing procedures, the segmentation process, and the classification techniques used for Covid-19 detection using deep learning, machine learning, and optimization algorithms are reviewed and categorized. After that, the dataset and the implementation tool which are utilized for Covid-19 detection works are analyzed and grouped. Finally, the performance metrics validation such as accuracy, recall, F1-score, NPV, precision, sensitivity, and specificity is carried out. The research gaps in the existing Covid-19 detection techniques are provided further as references to aid in future works.

As a classical rule-based modeling approach, belief rule base (BRB) expert system can integrate expert knowledge and possesses good interpretability. BRB with attribute reliability (BRB-r), built upon BRB, provides an effective way to deal with the problems of model reliability and environmental disturbances. Moreover, robustness is an important measure of perturbation resistance, and a robust BRB-r can remain reliable and stable in various environments. Therefore, to improve the model's ability to resist perturbations and enhance the model's adaptability, a new generic BRB with attribute reliability (G-BRB-r) is developed. Specifically, the robustness of BRB-r is analyzed in this paper to explore the change of BRB-r robustness under different perturbations. In addition, combining the effects of different factors on robustness, the construction criteria and constraints of robust BRB-r are given to guide modeling. Then, considering the effects of attribute reliability and robustness on modeling performance, a new generic BRB with attribute reliability is developed. Finally, the effectiveness and adaptability of the proposed method are demonstrated through a case study for health state assessment of the aerospace relay.

Machine learning algorithms treat credit risk prediction as a binary classification problem. However, two-way decisions with a single threshold force to make either a default or non-default decision may be inappropriate. To reduce the risk of decision errors, this study introduces three-way decisions and proposes a sequential three-way decision model with automatic threshold learning to evaluate credit risk. Initially, the model uses the loan amount and interest to determine the decision loss of the three-way decision, assigning distinct decision thresholds to different samples. Subsequently, the model employs decision cost and information gain to formulate an objective for threshold optimisation. Finally, the model continuously optimises the classification process by using the outcomes of certain decisions as supplementary information. In addition, to validate our model, we conduct comparative experiments with various methods on a real credit dataset from a Chinese bank. The results indicate that the model not only enhances classification performance across several metrics but also assists financial institutions in reducing decision error costs.

Social responsibility is a key factor for organizations to achieve sustainable success in the modern competitive market. This study proposes a hybrid VIKOR method to evaluate textile suppliers based on their social performance under uncertain and multi-objective conditions. The method can handle fuzzy, stochastic, and interval data simultaneously. The social criteria for the evaluation are derived from the literature review, the SA8000 standards, and the United Nations’ recommendations. Some of the criteria are also aligned with the World Bank’s Social Responsibility Diamond Model and the United Nations’ Sustainable Development Goals. Moreover, this study presents a fuzzy mathematical model for fabric purchasing that incorporates social criteria and the quality level into the optimization process. A goal programming method is developed based on the mathematical properties of the multi-objective model. A numerical study is conducted in the textile industry to demonstrate the efficiency and effectiveness of the proposed approaches. A comprehensive sensitivity analysis has been performed to investigate the behavior of the presented mathematical model under different conditions, and the results have been discussed concerning the insights for managers and stakeholders in the textile industry. The proposed model demonstrates that: 1) Customer demand and fabric orders have a direct relationship with increasing sales. 2) The fabric unit price has a direct impact on the quality value and requires cost control policies or pricing negotiations with suppliers. 3) Improving supplier and customer relations and formulating pricing consistent with social value are among the most important issues for the success of the textile and clothing industry. The best-fitting line successfully explains the variability of social performance and customer demand with an accuracy of 99.35 %.

In surface defect detection, some regions remain ambiguous and cannot be distinctly classified as abnormal or normal. This challenge is exacerbated by subjective factors, including workers’ emotional fluctuations and judgment variability, resulting in noisy labels that lead to false positives and missed detections. Current methods depend on additional labels, such as clean and multi-labels, which are both time-consuming and labor-intensive. To address this, we utilize Rough Set theory and Bayesian neural networks to learn a trustworthy model from noisy labels for Surface Defect Detection. Our approach features a novel pixel-level representation of suspicious areas using lower and upper approximations, and a novel loss function that emphasizes both precision and recall. The Pluggable Spatially Bayesian Module (PSBM) we developed enhances probabilistic segmentation, effectively capturing uncertainty without requiring extra labels or architectural modifications. Additionally, we have devised a ‘defect discrimination confidence’ metric to better quantify uncertainty and assist in product quality grading. Without the need for extra labeling, our method significantly outperforms state-of-the-art techniques across three types of datasets and enhances seven types of classic networks as a pluggable module, without compromising real-time computing performance. For further details and implementation, our code is accessible at https://github.com/ntongzhi/RoughSet-BNNs.

Thoracic diseases are a major source of mortality, often requiring diagnosis from plain chest X-rays. However, differentiating between complex conditions based on subtle radiographic patterns poses challenges even for experts. Recently, deep learning methods have shown promise in automating thoracic disease detection from chest radiographs. Many existing approaches focus on the diseased organs in the radiographs by utilizing spatial regions that contribute significantly to the model’s prediction. Expert radiologists, on the other hand, first identify the prominent region before determining whether those regions are abnormal or not. Therefore, incorporating localization information through deep learning models could result in significant improvements in automatic disease classification. Motivated by this, we have proposed a generalized weakly supervised Confidence-Aware Probabilistic Class Activation Map (CAPCAM) classification model that localizes anomalies for thoracic disease. The CAPCAM used CX-Ultranet as the backbone with the combination of Confidence Aware Network (CAN) and Anomaly Detection Network (ADN) without having any localization labeling. This learning from the backbone helps the model to utilize all components of the feature extracted and, therefore eliminating the need to train them individually reducing the time taken. We have experimentally shown that the proposed CAPCAM method sets a new state-of-the-art benchmark by achieving accuracy in terms of Intersection of bounding box (IoBB) in the range of 85 % - 94 %, and Dice scores in the range of 88 %-90 % for all thirteen diseases on two publicly available large-scale CXR datasets–NIH, Stanford and CheXpert. Testing across different noise levels and different levels of blurred level assessed real-world viability. We have also added a layer of explainability to show how the image is processed. This study demonstrates deep learning’s potential to augment radiologists’ decision-making by providing fast, accurate automated aids for thoracic disease diagnosis. The proposed CAPCAM model could be readily translatable to improve clinical workflows.

The Linear Complementarity Problem (LCP) offers a comprehensive modeling framework for addressing a wide range of optimization problems. In many real-world applications, finding an LCP solution with a sparse structure is often necessary. To address this problem, we introduce an innovative global optimization framework named the Particle Dynamical System Algorithm (PDSA), which consists of two components. The first component is a dynamical system (DS) inspired by the Absolute Value Equation (AVE), proven to have equilibria corresponding to LCP solutions, with additional relaxing regulators that enhance coverage rate and stability. The second component is an Adaptive Oscillated Particle Swarm Optimization (AOPSO) designed to globally enhance sparsity in LCP solutions, addressing the complexities posed by non-convex and non-smooth regulation models. Within this framework, the DS achieves optimality, while the AOPSO promotes solution sparsity. We compared our proposed DS with relaxing regulators to two classic efficient DSs, fully validating the effectiveness of our approach and underscoring the significant role of the introduced relaxing regulators in improving the convergence rate. Our newly developed variant of PSO, AOPSO, was compared with three classic and state-of-the-art variants on fourteen benchmark functions, demonstrating its competitive performance. Finally, we performed experiments on seven test examples and an application in portfolio selection, showing that the proposed PDSA algorithm surpasses other competitors in finding sparse LCP solutions.

Given the continual evolution of new network attack methodologies, defenders face the imperative of constantly upgrading security defenses. Current security technologies, albeit effective against known threats, often fall short in handling the intricacies of diverse and novel attacks. Artificial immunity-based network anomaly detection offers a promising avenue by dynamically adapting to evolving threats. However, prevailing algorithms in this domain suffer from low detection rates, limited adaptability, and extended detector generation times. This study aims to tackle these challenges by introducing a high-efficiency network anomaly detection framework, emphasizing both high-dimensional feature selection and adaptive detector generation. Our approach begins with an enhanced dual-module hybrid high-dimensional feature selection method, leveraging evolutionary principles. Furthermore, we introduce a self-sample clustering algorithm based on fuzzy clustering during the tolerance stage, enhancing detector tolerance efficiency. Additionally, an adaptive detector generation scheme is devised. It divides the non-boundary sub-population based on non-self differences and evolution, while employing the red fox optimization algorithm in the boundary region. This adaptive approach dynamically adjusts detector positions and radii to derive optimal detectors. Through comprehensive validation using well-established IoT and network anomaly datasets, our proposed artificial immunity-based IoT intrusion detection framework exhibits superior performance. It achieves higher classification accuracy and lower error rates compared to current state-of-the-art machine learning and artificial immunity algorithms.

The strong noise often masks the fault characteristics of equipment, which reduces the accuracy of fault diagnosis and even leads to the inability of intelligent fault diagnosis algorithms to be applied in industrial environments. This has always been a challenge in the field of mechanical fault diagnosis. As known that equipment failure results from the continuous degradation of the equipment’s state, with the failure state evolving from the healthy state. Considering that both healthy signals and fault signals contain similar noise, this paper proposes a Reference Learning Network (RLNet) model. The model aims to enhance the distinguishing features between healthy and faulty samples through reference units, thereby eliminating the influence of noise on feature distribution. Firstly, the impact of variable speed on the model’s robustness is mitigated using the computed order tracking method. Then, the difference features between healthy samples and a class of fault samples are extracted through the binary classification reference learning unit (RLU). Next, the extracted differential features are used to train the state classifier. Finally, membership weights are employed to effectively combine the feature recognition results, reducing the influence of fault features from mismatched RLUs. The robustness and superiority of the proposed method were verified by comparing it with five other intelligent fault diagnosis methods on the gear and bearing datasets. RLNet is of great significance for the engineering application of intelligent fault diagnosis methods in industrial noise environments.