machine learning

Today, smartphones are prevalent for personal and corporate use and have become the new personal computer due to their portability, ease of use, and functionality (such as video conferencing, Internet browsing, e-mail, continuous wireless and data connectivity, worldwide map location services, and countless mobile applications such as banking applications). On the other hand, we store many sensitive and private information daily on smart devices. This information is of interest to malicious writers who are developing malware to steal information from mobile devices. Unfortunately, the open source and widespread adoption of the Android operating system has made it the most targeted of the four popular mobile platforms by malware writers. Many researchers have tried to identify malware using program signatures, which have been successful to some extent. However, the signature cannot effectively identify new and unknown malware. For this reason, in this article, we propose a method that designs a machine-learning model for Android malware detection based on the properties of Permissions, Intents APKs. In this study, we evaluated more than 25,000 Android samples belonging to malware and trusted samples. Experimental results show the effectiveness of the proposed method by obtaining 96.27% accuracy.

There is an urgent need for precise and trustworthy models to forecast device behavior and evaluate vulnerabilities as a result of the Internet of Things' (IoT) explosive growth. By assessing the effectiveness of several machine learning algorithms logistic regression, decision trees, random forests, Naïve Bayes, and KNN on two popular IoT devices Alexa and Google Home Mini this study seeks to enhance IoT device behavior forecasting. Our results show that Naïve Bayes and random forest models are more accurate and efficient than other algorithms at predicting device behavior. These findings demonstrate how important algorithm selection is for maximizing the performance of IoT systems. The study also emphasizes the usefulness of precise device behavior prediction for practical uses such as industrial control systems, home automation, and medical monitoring. For example, accurate forecasts can improve decision-making in crucial situations, facilitate more seamless automation, and stop system failures. In addition to adding to the expanding corpus of research on IoT data analysis, this study establishes the foundation for the creation of increasingly sophisticated machine learning models that can manage the intricate and ever-changing nature of IoT ecosystems. Future studies should concentrate on increasing the dataset's diversity to encompass a wider range of IoT environments and devices and enhancing the model's adaptability to changing IoT environments.

This research investigates the impact of artificial intelligence techniques, including machine learning, deep learning, and neural networks, on optimizing production processes in the electronics industry, focusing on companies based in Tehran. The research sample consisted of 243 managers and experts related to supply chain management within these companies. Using Partial Least Squares Structural Equation Modeling (PLS-SEM), the results indicated that each of these techniques positively and directly influences demand forecasting, inventory management, quality control, and cost reduction in production. The findings also highlight the importance of integrating these techniques to achieve synergistic effects and enhance overall production process performance. From a theoretical perspective, this study contributes to expanding the existing literature on the application of artificial intelligence in manufacturing and presents new models for examining the interactions of these techniques. On the other hand, the results are practical for managers in the electronics industry and other manufacturing sectors, enabling them to leverage artificial intelligence tools for process optimization, productivity enhancement, and cost reduction. Ultimately, this research demonstrates that artificial intelligence can act as a transformative factor in the manufacturing industry, guiding organizations toward sustainable competitiveness.

In the light of growing frequency of natural disasters, social networking sites are now used for polling perception and evaluating governmental performance. The aim of this study is to examine the effects of negative and positive social media discussions to government responses with the floods disaster of 2010 in Pakistan. This study, being a sentiment analysis of tweets that involves the Pakistan Flood 2010 and Disaster Relief hashtags only, classified public responses as positive, negative, or neutral. The sentiments are transformed into actionable insights using the Enhanced Hybrid Dark Social Analytical Framework (EHDSAF) technique across different areas. The study advances knowledge of how public sentiment shapes government responses by showing that negativity correlates with slower response and revised policies. The majority of the tweets analyzed were neutral (45%), followed by positive (35%), and negative (20%). Negative sentiment tends to be concentrated during the peak crisis period. Higher negative sentiment, particularly in big cities correlates with more immediate and substantial government interventions, indicated by a strong correlation of 0.65. The Pearson correlation coefficient calculated as 0.68, suggests a strong relationship between public sentiment and response. The study therefore establishes social media as an accountability forum that provides real-time feedback to government agencies in the course of calamity management. This paper highlights the effectiveness of using sentiment analysis to update the approach by which disasters are responded to, as well as improve the perception of the public towards government endeavors.

The digital era has introduced mental health challenges, especially for youth. Despite increasing awareness, comprehensive analyses of these challenges remain limited. This study collects and examines the prevalence of 15 key mental health challenges related to digital engagement, based on a sample of 555 participants. The prevalence of these challenges varied, with pressures related to parenting, hoarding, and inappropriate content being the most common, affecting 60.13%, 52.76%, and 45.39% of the participants, respectively. The research also highlights gender and age differences, noting that males report higher levels of issues like FOMO and Nomophobia compared to females. Adults (18+) face more severe challenges, such as memory decline, while younger individuals report fewer problems. Correlation analysis revealed significant relationships between several mental health challenges, such as Nomophobia and TAD (r = 0.68) and FOMO and TAD (r = 0.50), indicating that individuals experiencing one challenge are likely to face others. A decision tree analysis was used to predict mental health challenges by examining the relationships between different mental health conditions, uncovering specific patterns and rules associated with the occurrence of these challenges. Additionally, cluster analysis in this study identified distinct population segments, with 21% of individuals falling into a cluster that experiences severe mental health challenges. The findings suggest that a significant portion of the population is at risk for severe mental health issues, highlighting the need for targeted interventions.

The rapid advancement of artificial intelligence (AI) offers new avenues for enhancing the identification, control, and autonomous operation of marine vehicles. This study investigates the application of AI techniques in maritime environments, focusing on object detection, navigation, and autonomous control to support safe and efficient operations in various marine conditions. Key objectives include evaluating machine learning models for identifying and tracking marine vehicles and the development of intelligent control algorithms that can adapt to dynamic oceanic settings. Methods involved training convolutional neural networks (CNNs) on datasets of marine images for object identification and using reinforcement learning (RL) algorithms to optimize the control systems of autonomous marine vehicles. Results demonstrate that CNN-based models achieve high accuracy in vehicle identification, even under challenging visual conditions such as low lighting or occlusion. At the same time, RL-driven control systems adapt effectively to complex, fluctuating marine environments. Simulated and real-world testing indicated that these AI techniques improve vessel maneuverability and response times, leading to more efficient and safer operations. In conclusion, this study highlights the potential of AI to revolutionize marine vehicle identification and control, with implications for enhanced security, efficiency, and sustainability in maritime operations. It is advisable to conduct additional research to improve these models, enabling their application across a wider range of marine environments.

In today's world, DDoS attacks are becoming more common and complex; thus, they constitute a great challenge for network security under the auspices of SDN. The research effort described here proposes an integrated hybrid model called "LDA-ML," which leverages some state-of-the-art machine learning methods: LDA, naive bayes, random forest, and logistic regression. We optimize the data analysis process by leveraging LDA for feature selection and dimensionality reduction, followed by a sequential application of the classifiers to exploit their strengths. Evaluated on the CICDDoS-2019 dataset, the proposed model has achieved an outstanding accuracy of 98.98%, indicating the efficacy of the model in correctly classifying benign versus attack traffic. All of the above underlines the robustness of the proposed LDA-ML model, pointing to great potential for its application to continuously improve cybersecurity strategies against DDoS threats in SDN architectures. This holistic approach offers improvements in detection, while it also enriches diagnostic insights-an important contribution to finding effective security solutions in increasingly dynamic network environments.

The energy sector encompasses essential processes such as the production, distribution, and consumption of energy. Traditionally, these processes have been managed through conventional networks, which often lead to issues such as process fluctuations, increased costs, and inefficiencies. However, the advent of Internet of Energy technology facilitates a transition from traditional to smart networks. In the Internet of Energy, the use of sensors results in the generation of large volumes of data. By employing machine learning to analyze this data, it becomes possible to make accurate predictions in the energy sector, which in turn supports effective decision-making for energy production and distribution. The objective of this study is to analyze data within the Internet of Energy using machine learning techniques, ultimately leading to the development of an artificial intelligence model capable of predicting energy consumption. Initially, previous models will be reviewed, and their outcomes will be compared and analyzed based on scores and evaluation metrics. Finally, a deep neural network model will be introduced, demonstrating an error rate of 0.3. The mean absolute error is reported as 0.4, and the mean square error is 0.3. Despite these advantages, there are also limitations to consider. The data involved in the analysis and prediction process must meet appropriate standards. The significant variability present in industrial processes adds complexity to the environment.

In this paper, a mmWave antenna is designed and optimized by machine learning to meet 5G requirements. A Surrogate based optimization approach is used to optimize this antenna. The goals of this optimization are antenna gain, bandwidth, and size. The bandwidth of the optimized antenna covers the 27.5 GHz- 38.9 GHz frequency band. This antenna has a wide 3 dB beam width with 90∘ and 178∘ beam width in the E-plane and H-plane, respectively. The realized gain of the antenna is 5 dB in 28 GHz frequency. One of the benefits of this antenna is Its compact size. The total dimension of the antenna is 5.4×5.6 mm2. In addition, an array of this antenna is proposed with 4 elements. The analog beamforming is achieved by means of this array. The 4-element array is simulated in an iPhone package model. Finally, an antenna with a wide beam, wide band, high gain, and small size is achieved within the Ka-band.

Structural reliability analysis (SRA) is associated with complex calculations and large number of simulations. In this paper, machine learning (ML) methods are integrated with SRA to reduce the overall intricacy and computational cost of direct SRA methods, such as the Monte Carlo simulation (MCS) method. An SRA is conducted in this paper on post-tensioned concrete members under the influence of prestress loss, and their reliability indices are obtained through the MCS method. The results of the SRA are used to create a database for data fitting of the ML algorithms. The algorithms are compared to find the most accurate ML model to be applied on the problem at hand. For the SRA, different stochastic parameters with specified probabilistic distributions are considered for the numerical models, and nonlinear dynamic analyses are conducted on them. Using the labeled data resulted from the SRA, five ML algorithms are compared; (i) linear regression, (ii) random forest, (iii) artificial neural network, (iv) k-nearest neighbors, (v) extreme gradient boosting. R-squared and root mean squared error are considered as the metrics used for the comparison of the ML models. Bayesian search is used for hyperparameter optimization of algorithms. The performance of the linear regression algorithm (R2=0.67 and RMSE=0.26) indicates that the SRA problems are highly nonlinear and linear algorithms cannot precisely map the relationships in data. However, the results show that extreme gradient boosting has the finest accuracy with R2=0.9 and RMSE=0.04. Additionally, its predicted values mostly have relative errors of less than ±30%.

Air pollution emerges as a formidable threat to both public health and environmental integrity, especially in regions undergoing rapid development. In this context, Gujarat, situated in Western India, grapples with escalating air quality degradation attributable to industrialization, vehicular emissions, and agricultural practices. The imperative to accurately forecast Air Quality Indices (AQIs) becomes paramount for the prompt implementation of mitigation measures, thereby safeguarding public well-being. This research delves into the utilization of Machine Learning (ML) algorithms, specifically Random Forest (RF) and XGBoost, to predict AQIs in Gujarat. Four pivotal parameters, PM2.5, PM10, SO2, and NOX, are scrutinized due to their substantial impact on air quality and inclusion in publicly available datasets. Remarkably, both RF and XGBoost models exhibit outstanding performance, surpassing 99% accuracy. This exceptional capability underscores the transformative potential of ML in addressing the complex challenges posed by air pollution. Leveraging the precise predictions of AQI values, these models can catalyze the development of robust early warning systems and guide policymaking endeavors directed at enhancing air quality. Notably, this investigation unveils PM2.5 and PM10 as primary culprits influencing AQI levels in Gujarat, underscoring the urgency for stringent emission control measures targeting these pollutants. Furthermore, the study sheds light on the significant impact of meteorological factors, especially maximum temperature, on AQI fluctuations, necessitating adaptive strategies to counteract climate change's repercussions on air quality.

Today, the scope of using the Internet of Things is growing by taking science and technology as the first place in human life, and as these networks get bigger, more data are exchanged. It performs high-speed data exchanges on the Internet and in a pre-defined network. The more the Internet of Things penetrates into people's lives, the more important data it transmits. This causes attackers to draw attention to these data, and Internet of Things network devices that have limited resources are exposed to attacks. With the complexity of hardware and software for the ease of human's use, naturally more intelligent attacks will happen, which is the reason of presenting many methods in this field. For this reason, in this article, we are going to discuss the most important methods used in intrusion detection systems based on deep learning and machine that can identify these interruptions. In this article, we have compared 46 articles from 2020 to 2024 based on the type of dataset used, the type of classification (binary or multi-class) and the accuracy rates obtained from each method, and we have been able to see a comprehensive overview for researchers who intend to work in IoT data security. According to the obtained results, if the proposed method is implemented in binary form, it can achieve better accuracy than multi-class.

The detection of defects is important in quality control in manufacturing. These defects raise the costs incurred by enterprises, compress the service life of simulated products, and result in the expansive destruction of resources, thereby significantly harming people and their safety. Defect detection and classification need to be feasted as unique problems associated with the field of artificial vision. We categorize the defects like electronic components, pipes, welded parts, textile materials, etc. We express artificial visual processing techniques aimed at comprehending the charged picture in a mathematical/analytical manner. Recent mainstream and deep-learning techniques in defect detection are studied with their features, stability, and weaknesses explained. We resume with a survey of textural defect detection based on statistical, structural, and other methods. We investigate the application of ultrasonic testing, filtering, deep learning, machine vision, and other technologies utilized for defect detection to offer a new classification. In addition, high precision, high positioning, fast detection, and small objects through examination are the biggest challenges in applying quality detection.