Advances in computational intelligence最新文献

Supervised learning will be a bottleneck for developing plant disease identification since it relies on learning from massive amounts of carefully labeled images, which is costly and time-consuming. On the contrary, self-supervised learning has succeeded in various image classification tasks; however, it has not been applied broadly in the plant disease analysis process. This work, therefore, studies the effectiveness of self-supervised learning using contrastive pre-training with SimCLR for plant disease image classification. We investigated unsupervised pre-training scenarios on unlabeled plant images across multiple architectures, including supervised fine-tuning on labeled samples. In addition, we explored the label efficiency of the self-supervised approach, acquired by fine-tuning the models on various fractions of labeled images. Our results demonstrated that the performance of self-supervised learning on plant disease became comparable to that of the supervised training approach.

In this paper, we have considered a non-linear mathematical model to study the chaotic situation, arising due to slow process of recruitment, leading to an increase in unemployment. We observed the effects on recruitment process due to delay and without delay. We have also studied the stability of equilibrium points with numerical examples to compare with analytical and theoretical results.

Threat surface area for the Internet of Things is calculated as the sum of security vulnerabilities or the weakness and gaps in protection efforts for the device, operating systems, associated software applications, and the local infrastructure. This aggregates all the known and unknown threats that can potentially expose the device, logs, data, and hosted applications. By reducing the exposed elements of the device surface, the device vulnerabilities can decrease the exposed threat surface area. This research presents a new framework first to map the devices in the ecosystem, measure the potential threat surface area from the exposure indicators for each layer and then determine the threat vectors for device compromise to calculate the maturity and severity levels. The authors propose new metrics to reassess and re-calculate the maturity and severity levels. Based on the new metrics, newly exposed threat surface elements provide a new security perspective beneficial for stakeholders involved in design, implementation, and security ecosystem of smart devices.

Local citation recommendation is a list of references that researchers need to cite based on a given context, so it could help researchers produce high-quality academic writing quickly and efficiently. However, existing citation recommendation methods only consider contextual content or author information, ignore the critical influence of historical citation information on citations, and learn the paper embedding at a coarse-grained level, resulting in lower-quality recommendations. To solve the above problems, we propose a novel two-stage citation recommendation model with multiple information fusion (MICR). The first stage is to enhance the target paper’s representation learning of the MICR model. To achieve the above goal, three encoders, which contain context information encoder, historical citation encoder, and author information encoder, are designed to learn rich representations of the target paper. The second stage is to select appropriate recommendation strategies for the target paper and candidate papers to achieve the goal of efficient citation recommendation. Experiments on two public citation datasets show that our model outperforms several competitive baseline methods on citation recommendation.

In the real number set, generalized intuitionistic fuzzy numbers (GIFNs) are an impressive number of fuzzy sets (FSs). GIFNs are very proficient in managing the decision-making problem data. Our aim of this paper is to develop a new ranking method for solving a multi-attribute decision-making (MADM) problem with GIFN data. Here, we have defined the possibility mean and standard deviation of GIFNs. Then, we have formulated the magnitude of membership and non-membership function of GIFNs. In the proposed MADM problem, the attribute values are expressed as GIFNs, which is a very workable environment for decision-making problems. Finally, a numerical example is analyzed to demonstrate the flexibility, applicability and universality of the proposed ranking method and MADM problem.

In this study, a decision support system (DSS) based on the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method was developed using MATLAB to select the best dental implant alternative. The first step involved conducting interviews with experts to identify the criteria for TOPSIS. In the second step, a database was structured for each dental implant brand distributed in the market for the last five years. In the third step, MATLAB code and Graphical User Interfaces (GUI) were created to execute TOPSIS. The user can also display the other four options with a graph on the GUI, including the ranking scores (C_i^*) for each option. The DSS was applied in two case studies. The MATLAB-based DSS tool has a compact, user-friendly interface, making it easy to adopt in implant selection decisions. The proposed DSS can be widely used in different applications in dental implant selection tasks.

Speech mood analysis is a challenging task with unclear optimal feature selection. The nature of the dataset, whether it is from an infant or adult, is crucial to consider. In this study, the characteristics of speech were investigated using Mel-frequency cepstral coefficients (MFCC) to analyze audio files. The CREMA-D dataset, which includes six different mood states (normal, angry, happy, sad, scared, and irritated), was employed to identify mood states from speech files. A mood classification system was proposed that integrates Support Vector Machines (SVM) and Long Short-Term Memory (LSTM) models to increase the number of labeled data in small datasets and improve classification accuracy.

A semi-supervised model was proposed in this study to improve the accuracy of speech mood classification systems. The approach was tested on a classification model that used SVM and LSTM, and it was found that the semi-supervised model outperforms both SVM and LSTM models, achieving a validation accuracy of 89.72%. This result surpasses the accuracy achieved by SVM and LSTM models alone. Moreover, the semi-supervised method was observed to accelerate the training process of the model. These outcomes illustrate the efficacy of the proposed model and its potential to enhance speech mood analysis techniques.

As the adoption of Industry 4.0 advances and the manufacturing process becomes increasingly digital, the Digital Twin (DT) will prove invaluable for testing and simulating new parameters and design variants. DT solutions build a 3D digital replica of the physical object allowing the managers to develop better products, detect physical issues sooner, and predict outcomes more accurately. In the past few years, Digital Twins (DTs) dramatically reduced the cost of developing new manufacturing approaches, improved efficiency, reduced waste, and minimized batch-to-batch variability. This paper aims to highlight the evolution of DTs, review its enabling technologies, identify challenges and opportunities for implementing DT in Industry 4.0, and examine its range of applications in manufacturing, including smart logistics and supply chain management. The paper also highlights some real examples of the application of DT in manufacturing.

In recent years, cyber security attacks have increased massively. This introduces the need to defend against such attacks. Cyber security threat intelligence has recently been introduced to secure systems against security attacks. Cyber security threat intelligence (CTI) should be fast, trustful, and protect the sender's identity to stop these attacks at the right time. Threat intelligence sharing is vitally important since it is considered an effective way to improve threat understanding. This leads to protecting the assets and preventing the attack vectors. However, there is a paradox between the privacy safeguard needs of threat intelligence sharing; the need to produce complete proper threat intelligence feeds to be shared with the community, and other challenges and needs that are not covered in the traditional CTI. This paper aims to study how Blockchain technology can be incorporated with the CTI to solve the current issues and challenges in the traditional CTI. We collected the latest contributions that use Blockchain to overcome the conventional CTI problems and compared them to raise the reader’s awareness about the different methods used. Also, we mentioned the uncovered areas for each paper to offer a wide range of details and information about different areas that need to be investigated. Furthermore, the prospect challenges of integrating the Blockchain and CTI are discussed.

Cloud computing technology is growing fast. It offers end-users flexibility, ease of use, agility, and more at a low cost. This expands the attack surface and factors, resulting in more attacks, vulnerabilities, and corruption. Traditional and old security controls are insufficient against new attacks and cybercrime. Technologies such as intrusion detection system (IDS), intrusion prevention system (IPS), Firewalls, Web Application Firewall (WAF), Next-Generation Firewall (NGFW), and endpoints are not enough, especially against a new generation of ransomware and hacking techniques. In addition to a slew of cloud computing options, such as software as a service (SaaS), it is challenging to manage and secure cloud technology. A new technique is needed to detect zero-day attacks related to ransomware, targeted attacks, or intruders. This paper presents our new technique for detecting zero-day ransomware attacks and intruders inside cloud technology. The proposed technique is based on a deception system based on honey files and tokens.