Phishing Dynamic Evolving Neural Fuzzy Framework for Online Detection “Zero-day” Phishing Email

The changeable structures and variability of email attacks render current email filtering solutions useless. Consequently, the need for new techniques to harden the protection of users' security and privacy becomes a necessity. The variety of email attacks, namely spam, damages networks' infrastructure and exposes users to new attack vectors daily. Spam is unsolicited email which targets users with different types of commercial messages or advertisements. Porn-related content that contains explicit material or commercials of exploited children is a major trend in these messages as well. The waste of network bandwidth due to the numerous number of spam messages sent and the requirement of complex hardware, software, network resources, and human power are other problems associated with these attacks. Recently, security researchers have noticed an increase in malicious content delivered by these messages, which arises security concerns due to their attack potential. More seriously, phishing attacks have been on the rise for the past couple of years. Phishing is the act of sending a forged e-mail to a recipient, falsely mimicking a legitimate establishment in an attempt to scam the recipient into divulging private information such as credit card numbers or bank account passwords (James, 2005). Recently phishing attacks have become a major concern to financial institutions and law enforcement due to the heavy monetary losses involved. According to a survey by Gartner group, in 2006 approximately 3.25 million victims were spoofed by phishing attacks and in 2007 the number increased by almost 1.3 million victims. Furthermore, in 2007, monetary losses, related to phishing attacks, were estimated by $3.2 billion. All the aforementioned concerns raise the need for new detection mechanisms to subvert email attacks in their various forms. Despite the abundance of applications available for phishing detection, unlike spam classification, there are only few studies that compare machine learning techniques in predicting phishing emails (Abu-Nimeh et al., 2007). We describe a new version of Bayesian Additive Regression Trees (BART) and apply it to phishing detection. A phishing dataset is constructed from 1409 raw phishing emails and 5152 legitimate emails, where 71 features (variables) are used in classifiers' training and testing. The variables consist of both textual and structural features that are extracted from raw emails. The performance of six classifiers, on this dataset, is compared using the area under the curve (AUC) (Huang & Ling, 2005). The classifiers include Logistic Regression (LR), Classification and Regression Trees (CART), Bayesian Additive Regression Trees (BART), Support Vector Machines (SVM), Random O pe n A cc es s D at ab as e w w w .in te ch w eb .o rg

No 182W excess in the Ontong Java Plateau source

Many models have been developed to predict hospital 30-day readmissions. In this study, AUTO-WEKA was utilized against datasets with differing class imbalances to identify potential algorithms for further exploration. The study identified two algorithms, KStar and IBk, along with attribute selection criteria with high levels of sensitivity, specificity, precision, and accuracy. The two algorithms were identified with datasets containing higher class imbalances. Lower class imbalance datasets were not able to produce algorithms with acceptable performance.

Reducing the impact of cyberthreats with robust data governance

Predication Corrosion rate is quantitative method by which the effectiveness of corrosion control and prevention techniques can be evaluated and provides the feedback to enable corrosion control and prevention methods to be optimized. In this paper, a novel Model to predict corrosion rate based on RBFNN was proposed. A model is produced from experimental work for one year and eighty four specimens were used through this work using anode with a high level of precision. Learning data was performed by using a 36 samples test with different Environment Resistivity (ER), Impressed Current (IC), Location of Anode (LA), Corrosion Current (CC) and Corrosion Rate (CR). The RBFNN model has five input nodes representing the (ER, IC, CC, LN, and SA), sixteen nodes at hidden layer and one output node representing corrosion rate (CR). Simulation test use 6 data samples taken from the experimental results to check the performance of the neural network on these data and shows the proposed model can be use successfully to predicate the corrosion rate.

Named entity detection and Parts-of-speech tagging are the key tasks for many\nNLP applications. Although the current state of the art methods achieved near\nperfection for long, formal, structured text there are hindrances in deploying\nthese models on memory-constrained devices such as mobile phones. Furthermore,\nthe performance of these models is degraded when they encounter short,\ninformal, and casual conversations. To overcome these difficulties, we present\nLiteMuL - a lightweight on-device sequence tagger that can efficiently process\nthe user conversations using a Multi-Task Learning (MTL) approach. To the best\nof our knowledge, the proposed model is the first on-device MTL neural model\nfor sequence tagging. Our LiteMuL model is about 2.39 MB in size and achieved\nan accuracy of 0.9433 (for NER), 0.9090 (for POS) on the CoNLL 2003 dataset.\nThe proposed LiteMuL not only outperforms the current state of the art results\nbut also surpasses the results of our proposed on-device task-specific models,\nwith accuracy gains of up to 11% and model-size reduction by 50%-56%. Our model\nis competitive with other MTL approaches for NER and POS tasks while outshines\nthem with a low memory footprint. We also evaluated our model on custom-curated\nuser conversations and observed impressive results.\n

The proliferation of artificial intelligence applications necessitates a clear understanding of the fundamental distinctions between Machine Learning (ML) and Deep Learning (DL) approaches. This study presents a systematic comparative analysis through a multi-dimensional evaluation framework. We analyzed 150 implementations across three domains (computer vision, natural language processing, and structured data analysis), evaluating performance metrics, resource utilization, and architectural complexities. Our findings reveal that while DL architectures achieve superior accuracy in complex pattern recognition tasks (mean improvement: 27.3%, p < 0.001), they require substantially higher computational resources (GPU utilization: 89.2% vs. 23.7% for ML). Traditional ML demonstrates notable advantages in scenarios with limited datasets (<10,000 samples), exhibiting 3.8x faster training times and a 72% lower memory footprint. To guide implementation decisions, we developed a quantitative decision matrix based on five critical parameters: data volume, computational constraints, problem complexity, interpretability requirements, and time sensitivity. The matrix achieved 91.4% accuracy in predicting the optimal approach across 50 independent test cases. This research provides empirical evidence for the trade-offs between ML and DL, offering practitioners a structured framework for algorithm selection while considering resource constraints and performance requirements.

Photovoltaic (PV)-based multi-product systems are the most effective strategies for optimizing energy consumption and enhancing energy production efficiency, which integrates multiple energy production processes within a single framework for maximizing the utilization of available energy. This research aims to provide an inclusive analysis of the PV-based multi-product system performance, which integrates electrolysis, and desalination units for simultaneous generation of power, hydrogen, and freshwater. Therefore, the optimized Extreme Gradient Boosting (XGBoost) model was carefully developed and trained to ensure high levels of accuracy and precision in predicting the power, fresh water, and hydrogen outputs of the system under varying operational conditions. This optimized XGBoost model is achieved through the hyperparameter tuning by the HenryGas Solubility Optimization Algorithm, which ensures high levels of accuracy and precision. The day-by-day analysis of power and hydrogen generation demonstrated that in winter, a minimum PV power of 6,200 kW and a hydrogen mass flow rate of 65 kg were witnessed due to low solar radiation and short daylight hours. In contrast, summer and spring registered the highest PV power levels of 8,000 kW and 7,500 kW, respectively, and hydrogen mass flow rates of 75 kg due to high solar radiation and prolonged daylight hours, hence enhancing system performance.

The increasing complexity of urban transportation systems and the growing volume of vehicles have made traffic congestion a persistent challenge in modern cities. Efficient traffic flow prediction is essential for mitigating congestion, improving road safety, optimizing traffic signal control, and enhancing overall transportation efficiency. In recent years, artificial intelligence (AI) has emerged as a transformative tool in the field of traffic management, offering sophisticated algorithms capable of modeling, analyzing, and predicting complex traffic patterns with high accuracy. The application of AI in traffic flow prediction leverages vast amounts of real-time and historical data to generate precise forecasts, supporting data-driven decision-making by urban planners and traffic control authorities. The prediction of traffic flow involves analyzing time-series data that exhibit nonlinear, dynamic, and often stochastic behavior. Traditional statistical models, such as autoregressive integrated moving average (ARIMA), have proven to be limited in handling the high dimensionality and variability inherent in traffic systems. In contrast, AI algorithms possess the capacity to learn and adapt from complex data inputs without the need for explicit programming, making them particularly suitable for traffic-related applications. AI algorithms used in traffic flow prediction can be broadly categorized into machine learning (ML) and deep learning (DL) approaches. Machine learning algorithms such as k-nearest neighbors (KNN), support vector machines (SVM), decision trees, and random forests have demonstrated effectiveness in short-term traffic prediction tasks. These algorithms are capable of identifying hidden patterns in traffic data and adjusting to changes in traffic behavior over time. Ensemble methods, which combine the strengths of multiple learning models, further enhance prediction accuracy and robustness. Deep learning algorithms, a subfield of AI inspired by the human brain’s neural architecture, have shown exceptional performance in capturing spatial-temporal dependencies in traffic data. Recurrent neural networks (RNNs), particularly long short-term memory (LSTM) networks and gated recurrent units (GRUs), are widely used for their ability to process sequential data and retain information over extended time intervals. Convolutional neural networks (CNNs) are employed to extract spatial features from traffic sensor data or road network imagery. Hybrid models that integrate CNNs with RNNs have achieved high levels of predictive precision by simultaneously learning spatial and temporal correlations. In addition to supervised learning methods, unsupervised and reinforcement learning techniques are also applied in traffic flow prediction. Clustering algorithms, such as k-means and DBSCAN, assist in identifying traffic patterns, while reinforcement learning models optimize adaptive traffic signal control systems by learning optimal actions through environmental interaction. This study explores the different types of AI algorithms used in traffic flow prediction, examining their theoretical foundations, structural differences, and practical applications. It aims to evaluate the comparative advantages of various algorithms in addressing the challenges of real-time traffic prediction in increasingly complex transportation networks.

We have developed a DNA aptamer-conjugated graphene field-effect transistor (GFET) biosensor platform to detect receptor-binding domain (RBD), nucleocapsid (N), and spike (S) proteins, as well as viral particles of original Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) coronavirus and its variants in saliva samples. The GFET biosensor is a label-free, rapid (≤20 min), ultrasensitive handheld wireless readout device. The limit of detection (LoD) and the limit of quantitation (LoQ) of the sensor are 1.28 and 3.89 plaque-forming units (PFU)/mL for S protein and 1.45 and 4.39 PFU/mL for N protein, respectively. Cognate spike proteins of major variants of concern (N501Y, D614G, Y453F, Omicron-B1.1.529) showed sensor response ≥40 mV from the control (aptamer alone) for fM to nM concentration range. The sensor response was significantly lower for viral particles and cognate proteins of Middle East Respiratory Syndrome (MERS) compared to SARS-CoV-2, indicating the specificity of the diagnostic platform for SARS-CoV-2 vs. MERS viral proteins. During the early phase of the pandemic, the GFET sensor response agreed with RT-PCR data for oral human samples, as determined by the negative percent agreement (NPA) and positive percent agreement (PPA). During the recent Delta/Omicron wave, the GFET sensor also reliably distinguished positive and negative clinical saliva samples. Although the sensitivity is lower during the later pandemic phase, the GFET-defined positivity rate is in statistically close alignment with the epidemiological population-scale data. Thus, the aptamer-based GFET biosensor has a high level of precision in clinically and epidemiologically significant SARS-CoV-2 variant detection. This universal pathogen-sensing platform is amenable for a broad range of public health applications and real-time environmental monitoring.

This article describes some of the experiences, challenges, lessons learned and success stories during the first two years of using a hybrid learning approach using laptops at the University of Ontario Institute of Technology. We present two different aspects of our experience: the use of technology in and outside of the classroom, and the necessary institutional support for our technology-based learning approach. Both aspects are presented as perceived directly by faculty. In both areas we discuss the approaches that have worked successfully as well as challenges.

We establish a fuzzy deep model called the fuzzy deep belief net (FDBN) based on fuzzy restricted Boltzmann machines (FRBMs) due to their excellent generative and discriminative properties. The learning procedure of an FDBN is divided into a pretraining phase and a subsequent fine-tuning phase. In the pretraining phase, a group of FRBMs is trained in a greedy layerwise way: the first FRBM is trained by original samples, and the average values of the left and right probabilities produced by its hidden units are treated as the training data for subsequent FRBMs. The resulting FDBN is either a generative or a discriminative model depending on the choice of training a generative or a discriminative type of FRBM on top. Then, a hybrid learning approach is proposed to fine-tune this novel fuzzy deep model: the well pretrained fuzzy parameters are first defuzzified, and the FDBN with defuzzified parameters is fine-tuned by the wake–sleep or stochastic gradient descent algorithm. This hybrid strategy not only avoids learning an intractable fuzzy neural network, but also greatly improves the classification capability of the FDBN. The experimental results on MNIST, NORB, and 15 Scene databases indicate that the FDBN with the hybrid learning approach can handle high-dimensional raw images directly. It inherits the fine nature of the FRBM and outperforms some state-of-the-art discriminative models in classification accuracy. Moreover, it shows better capability of robustness than a deep belief net when encountering noisy data.

This study proposes a novel hybrid learning approach for developing a visual path-following algorithm for industrial robots. The process involves three steps: data collection from a simulation environment, network training, and testing on a real robot. The actor network is trained using supervised learning for 500 epochs. A semitrained network is then obtained at the $250^{th}$ epoch. This network is further trained for another 250 epochs using reinforcement learning methods within the simulation environment. Networks trained with supervised learning (500 epochs) and the proposed hybrid learning method (250 epochs each of supervised and reinforcement learning) are compared. The hybrid learning approach achieves a significantly lower average error (30.9 mm) compared with supervised learning (39.3 mm) on real-world images. Additionally, the hybrid approach exhibits faster processing times (31.7 s) compared with supervised learning (35.0 s). The proposed method is implemented on a KUKA Agilus KR6 R900 six-axis robot, demonstrating its effectiveness. Furthermore, the hybrid approach reduces the total power consumption of the robot’s motors compared with the supervised learning method. These results suggest that the hybrid learning approach offers a more effective and efficient solution for visual path following in industrial robots compared with traditional supervised learning.

The Domain Name System (DNS) protocol essentially translates domain names to IP addresses, enabling browsers to load and utilize Internet resources. Despite its major role, DNS is vulnerable to various security loopholes that attackers have continually abused. Therefore, delivering secure DNS traffic has become challenging since attackers use advanced and fast malicious information-stealing approaches. To overcome DNS vulnerabilities, the DNS over HTTPS (DoH) protocol was introduced to improve the security of the DNS protocol by encrypting the DNS traffic and communicating it over a covert network channel. This paper proposes a lightweight, double-stage scheme to identify malicious DoH traffic using a hybrid learning approach. The system comprises two layers. At the first layer, the traffic is examined using random fine trees (RF) and identified as DoH traffic or non-DoH traffic. At the second layer, the DoH traffic is further investigated using Adaboost trees (ADT) and identified as benign DoH or malicious DoH. Specifically, the proposed system is lightweight since it works with the least number of features (using only six out of thirty-three features) selected using principal component analysis (PCA) and minimizes the number of samples produced using a random under-sampling (RUS) approach. The experiential evaluation reported a high-performance system with a predictive accuracy of 99.4% and 100% and a predictive overhead of 0.83 µs and 2.27 µs for layer one and layer two, respectively. Hence, the reported results are superior and surpass existing models, given that our proposed model uses only 18% of the feature set and 17% of the sample set, distributed in balanced classes.

This study aims to produce learning materials for writing scientific articles for foreign students using the hybrid learning approach. This approach was chosen by utilizing the advantages of offline classes combined with online classes. This approach is in accordance with the characteristics of students as millennials. The development research method used consists of product design, product design, production, and product testing. The end result is learning materials to write scientific articles that are packaged in the form of modules, starting from the debriefing of concepts (f2f), planning of writing (f2f), writing a draft (ol), and editing the draft (f2f) writing process-genre based scientific articles with a hybrid learning approach for foreign students