Traditional software defect prediction (SDP) models mainly rely on software metrics or handcrafted structural features derived from dependency graphs. While network embedding techniques have recently been introduced, existing approaches often fail to fully capture higher-order and multi-hop dependencies among software components, limiting their ability to represent complex structural relationships within software systems. This study proposes a call graph-based embedding framework for software defect prediction that systematically captures structural dependencies among classes and integrates them with conventional software metrics. The goal is to improve predictive performance while preserving meaningful structural information that is often overlooked in traditional metric-based and graph-feature approaches. Ten real-world software projects were analyzed. For each project, call graphs were constructed and processed using six network embedding algorithms to generate 32-and 128-dimensional representations. A total of 550 models per project were developed using 25 dataset configurations and 22 classifiers. Performance was evaluated using Accuracy and AUC, with statistical significance assessed through non-parametric tests and effect size analysis. A cost-effectiveness analysis was conducted to determine practical deployment thresholds. The combination of call graph embeddings and software metrics consistently outperformed standalone approaches. The best-performing model (SVM with RBF kernel and GraRep embeddings) achieved a mean AUC of 70.73%, improving AUC by 5.2% over metric-only models and 12.8% over embedding-only models. Cost analysis indicates effectiveness for projects with lower defect densities. Call graph-based embeddings enhance SDP performance while providing actionable guidance for cost-effective adoption.
Software Fault Prediction (SFP) is the process of predicting fault-prone software constructs during the initial phases of software development. Deep Neural Networks (DNN) have been hugely successful in the field of computer vision, audio, etc., where the input data is correlated spatially and temporally. In contrast, SFP operates on tabular data, rows of software metrics that lack the inherent spatial structure exploited by convolutional architectures in image domains. The authors have tried to remodel the most successful Convolutional Neural Network (CNN) for tabular data. A novel framework is proposed employing a tree-based feature selection technique, LSelect, to find the most significant features and an adaptive 1-dimensional Convolutional Neural Network (ACNN) for the classification task, which selects an optimal learning rate automatically. ACNN converts the tabular data (1-D) into 2-D using adaptive pooling layers, thereby forming an image from 1-D data. The framework classification results (Area under Curve) are compared with nine state-of-the-art algorithms, such as XGBoost, LightGBM, etc., and performance is validated using the Bayesian Signed Rank Test. It is found that the proposed framework performs comparably with the state-of-the-art methods with reduced model complexity. Also, the LSelect feature selection technique improves average model performance by 1.3%.
Micro-expressions are brief, involuntary facial movements that reveal concealed emotions. While Convolutional Neural Networks (CNNs) are effective for recognizing such expressions, they are often computationally intensive and memory heavy. To address this, a Feature-Pruned and Weighted Convolutional Neural Network (FPW-CNN) is proposed to compress and accelerate performance. This model prunes filters and selects only the most relevant responses, reducing memory usage and computational load. Instead of stacking outputs, convolutional layers are flattened. Experiments on five spontaneous micro-expression datasets—SMIC, CASME, CASME II, CAS(ME)2, and SAMM—show that FPW-CNN achieves accuracies of 99.75%, 99.93%, 99.95%, 99.94%, and 99.90%, respectively. The model also reduces FLOPs by over 41% and maintains over 95% accuracy at 128×128 resolution, demonstrating robustness. These results confirm that FPW-CNN outperforms recent CNN and transformer models, making it suitable for real-time micro-expression recognition on resource-constrained devices.
The Internet of Things (IoT) with its pervasive connectivity imposes substantial challenges regarding privacy, security, and trust. Traditional security measures, such as cryptography and access control, often prove inadequate in the dynamic and resource-constrained IoT environment. The decentralized nature of IoT networks, coupled with the heterogeneity of devices, demands innovative security solutions. Trust is essential in human interaction and plays a crucial role in reducing challenges and security risks within IoT networks. This article focuses on trust-based methods for securing IoT systems, particularly within the context of Wireless Sensor Networks (WSNs). Trust-based approaches offer advantages such as lower computational complexity, leading to improved power management compared to traditional authentication and encryption methods. This article examines the latest trust models, as well as machine learning and deep learning algorithms specifically designed and implemented to enhance the security and reliability of trust-based WSN-assisted IoT applications.
Artificial intelligence (AI) has rapidly evolved into an essential tool in biomedical imaging and scientific publishing, facilitating efficient data processing, hypothesis generation, and image reconstruction through advanced computational models such as variational autoencoders, generative adversarial networks, and diffusion models. Comprehensive editorial procedures emphasize author accountability and the disclosure of AI usage to ensure scientific accuracy amid the increasing use of AI-assisted writing. Despite considerable advancements, difficulties persist that result in false claims due to inadequate performance measurements, highlighting the necessity for robust evaluation systems. AI methodologies employing diverse techniques offer an enhanced integration of heterogeneous medical data, thus augmenting diagnostic accuracy across many different fields. Global regulatory frameworks exhibit significant variation yet align on objectives for transparency, risk management, and lifecycle governance to facilitate the utilization of AI medical devices.
The Internet of Things (IoT) with its pervasive connectivity imposes substantial challenges regarding privacy, security, and trust. Traditional security measures, such as cryptography and access control, often prove inadequate in the dynamic and resource-constrained IoT environment. The decentralized nature of IoT networks, coupled with the heterogeneity of devices, demands innovative security solutions. Trust is essential in human interaction and plays a crucial role in reducing challenges and security risks within IoT networks. This article focuses on trust-based methods for securing IoT systems, particularly within the context of Wireless Sensor Networks (WSNs). Trust-based approaches offer advantages such as lower computational complexity, leading to improved power management compared to traditional authentication and encryption methods. This article examines the latest trust models, as well as machine learning and deep learning algorithms specifically designed and implemented to enhance the security and reliability of trust-based WSN-assisted IoT applications.
Artificial intelligence (AI) has rapidly evolved into an essential tool in biomedical imaging and scientific publishing, facilitating efficient data processing, hypothesis generation, and image reconstruction through advanced computational models such as variational autoencoders, generative adversarial networks, and diffusion models. Comprehensive editorial procedures emphasize author accountability and the disclosure of AI usage to ensure scientific accuracy amid the increasing use of AI-assisted writing. Despite considerable advancements, difficulties persist that result in false claims due to inadequate performance measurements, highlighting the necessity for robust evaluation systems. AI methodologies employing diverse techniques offer an enhanced integration of heterogeneous medical data, thus augmenting diagnostic accuracy across many different fields. Global regulatory frameworks exhibit significant variation yet align on objectives for transparency, risk management, and lifecycle governance to facilitate the utilization of AI medical devices.
The adoption of privacy-preserving techniques in healthcare is significant, especially while handling sensitive medical information. Traditional machine learning approaches raise significant concern regarding privacy, regulations, and data protection. Federated learning has emerged as an effective machine learning approach that enables a group of local models to collaboratively train the global model by sharing their updates instead of sharing the sensitive medical data. Nevertheless, a significant issue with federated learning is its vulnerability to various attacks, including model corruption and data tampering. The authors propose a methodology for developing a secure and privacy-safeguarded collaborative learning model by integrating zero knowledge proof (ZKP) with federated learning (FL). The proposed RS-ZKP methodology utilizes Pedersen commitments within ZKP to verify feature importance, ensuring that they fall within specified bounds without disclosing the actual values. The methodology is validated on two benchmark datasets using metrics accuracy, precision, recall, and F1 score.
This paper proposes a novel approach for collision avoidance during lane changes using Model Predictive Control (MPC). The proposed method integrates real-time trajectory planning with dynamic vehicle modeling to predict and optimize the vehicle's motion over a finite time horizon. The paper presents the fundamental principles of MPC, its integration with vehicle dynamics, and its application to real-time control. Simulation results demonstrate the effectiveness of MPC in optimizing trajectory planning and ensuring safety under various traffic scenarios. This paper provides a comprehensive comparison of MPC with other control models such as Proportional-Integral-Derivative (PID) control, Rule-Based Control (RBC), and Reinforcement Learning (RL)-based approaches. Simulation results demonstrate the effectiveness of the proposed method in a variety of traffic scenarios, including high-density and mixed-traffic environments. Experimental results highlight the relative performance of these models under simulated environments in MATLAB.
The adoption of privacy-preserving techniques in healthcare is significant, especially while handling sensitive medical information. Traditional machine learning approaches raise significant concern regarding privacy, regulations, and data protection. Federated learning has emerged as an effective machine learning approach that enables a group of local models to collaboratively train the global model by sharing their updates instead of sharing the sensitive medical data. Nevertheless, a significant issue with federated learning is its vulnerability to various attacks, including model corruption and data tampering. The authors propose a methodology for developing a secure and privacy-safeguarded collaborative learning model by integrating zero knowledge proof (ZKP) with federated learning (FL). The proposed RS-ZKP methodology utilizes Pedersen commitments within ZKP to verify feature importance, ensuring that they fall within specified bounds without disclosing the actual values. The methodology is validated on two benchmark datasets using metrics accuracy, precision, recall, and F1 score.
The primary objective of this work is to identify the scope of requirements for existing software projects by considering the source code metrics. In this work, the authors used Deep Q Network (DQN) for refactoring prediction, for which three open-source real-time projects, Antlr4, Junit, and Oryx, are taken into consideration. Different code metrics like Cyclomatic Complexity, Lines of Code, Coupling between Objects Lack of Cohesion are taken as input parameters. Based on the experimental analysis, five different types of refactoring predictions are performed, and the requirement of Extract Method and Move Class is higher as compared to other techniques. JUnit framework identifies a high scope of refactoring for which the performance evaluation metrics values are with an accuracy 97%, recall of .96, f-measure of 0.96, precison of 0.96. Q-Learning can be considered as one of the best techniques for identifying the need for code alteration by considering the probable values that resemble the Q-Value used for analysing the maintainability index.
Micro-expressions are brief, involuntary facial movements that reveal concealed emotions. While Convolutional Neural Networks (CNNs) are effective for recognizing such expressions, they are often computationally intensive and memory heavy. To address this, a Feature-Pruned and Weighted Convolutional Neural Network (FPW-CNN) is proposed to compress and accelerate performance. This model prunes filters and selects only the most relevant responses, reducing memory usage and computational load. Instead of stacking outputs, convolutional layers are flattened. Experiments on five spontaneous micro-expression datasets-SMIC, CASME, CASME II, CAS(ME)2, and SAMM-show that FPW-CNN achieves accuracies of 99.75%, 99.93%, 99.95%, 99.94%, and 99.90%, respectively. The model also reduces FLOPs by over 41% and maintains over 95% accuracy at 128x128 resolution, demonstrating robustness. These results confirm that FPW-CNN outperforms recent CNN and transformer models, making it suitable for real-time micro-expression recognition on resource-constrained devices.
Robotic manipulation is a challenging research area, particularly in developing policies that generalize across diverse real-world scenarios. While real-world data can be slow and unsafe, simulations offer a safer and faster alternative. However, training in simulations can still be time-consuming, hindering rapid model iteration. This paper explores using graphics processing unit (GPU) acceleration to speed up training for robotic manipulation tasks in simulations. By comparing GPU and CPU performance, we demonstrate a significant reduction in training time. The findings show that GPU hardware enhances policy development efficiency, accelerating research and applications, including sim-to-real transfer. Additionally, it broadens exploration of state and action spaces, providing agents with a diverse range of training experiences. A simulation benchmark was also created to test GPU acceleration, detailing task selection, environment setup, and performance measurement. This benchmark forms the basis for evaluating the speedup achieved by GPUs in training robotic manipulation models.
The inherent delivery risks of crowdfunding campaigns force crowdfunding platforms to take risk disclosure measures to alleviate information asymmetry between creators and crowdfunders. However, creators might perceive such risk disclosures as a threat to crowdfunding success. Based on the language expectancy theory, this study aims to examine how the contextual characteristics of risk disclosure affect crowdfunding performance. By separating the project description and R&C sections of 21,287 projects on Kickstarter from 2009 to March 2022, we find that the two-sided persuasion topics extracted by LDA model from the R&C section have positive impacts on the funding success. This result indicates that providing two-sided persuasion content in the R&C section can increase the credibility of crowdfunding project narratives and thus the crowdfunding performance. Furthermore, by incorporating the R&C text into the prediction models, the prediction accuracy of funding support is improved by 0.9% from 83.9%, and the results have been proven to be robust through cross-validation experiments.
This paper presents a systematic literature review on Intelligent Process Automation (IPA), focusing on its role in modernizing business operations. It explores the integration of AI, including machine learning and robotic process automation, and their combined impact on improving efficiency and supporting human-centric tasks. The review examines the synergy between AI and human expertise, highlighting how this collaboration enhances decision-making, reduces errors, and speeds up task execution. It also investigates how digital assistants within the IPA framework redefine operational processes, supported by case studies and empirical research. The paper addresses challenges and ethical concerns related to AI integration, offering insights on risk management and ethical AI use. In conclusion, it emphasizes the paradigm shift created by AI-driven IPA, optimizing business processes and focusing on cognitive tasks, and contributing to the ongoing discussion about AI's transformative impact on human-machine collaboration.
This article examines the performance of cooperative NOMA systems in a massive MIMO configuration inside the sixth-generation (6G) network, including scenarios with and without reconfigurable intelligent surfaces (RIS). The emphasis is on comprehending the effects of implementing static, dynamic, and intelligent RIS on the capacity of the cooperative NOMA system with differing user quantities. The study examines how different user loads and RIS densities affect average latency and effective area spectral efficiency (EASE). Integration of a proposal system in a novel way with water-filling in the multi-user channel logarithm to enhance capacity, average latency, and EASE for diverse user densities. The results show that RIS significantly improves capacity, average latency, and EASE, especially with intelligent RIS. All scenarios, in conjunction with the suggested algorithm, markedly improve network performance, particularly under conditions of increased user demand. Additionally, it enhances the system's capability and spectrum efficiency, especially when deployment numbers rise.
This article examines the performance of cooperative NOMA systems in a massive MIMO configuration inside the sixth-generation (6G) network, including scenarios with and without reconfigurable intelligent surfaces (RIS). The emphasis is on comprehending the effects of implementing static, dynamic, and intelligent RIS on the capacity of the cooperative NOMA system with differing user quantities. The study examines how different user loads and RIS densities affect average latency and effective area spectral efficiency (EASE). Integration of a proposal system in a novel way with water-filling in the multi-user channel logarithm to enhance capacity, average latency, and EASE for diverse user densities. The results show that RIS significantly improves capacity, average latency, and EASE, especially with intelligent RIS. All scenarios, in conjunction with the suggested algorithm, markedly improve network performance, particularly under conditions of increased user demand. Additionally, it enhances the system's capability and spectrum efficiency, especially when deployment numbers rise.
This work addresses a research gap in digital marketing by attempting to compare the effort in achieving the best organic search engine ranking with the effort in providing the best user experience in web navigation. The objective is to validate companies' efforts in the digital world, and the study is focused on the toy sector in Spain, specifically on the Google search engine, measuring the user experience in web browsing through neuromarketing biometrics. The top 30 results for each Google search were collected for the 638 keywords related to toys in Spain. Subsequently, the three best-positioned websites for the Google search results were determined, and their user experience was measured using neuromarketing biometrics, triangulated with qualitative research. This approach allows for contrasting brand authority in the digital world (visibility in a search) with the user experience in navigation (trust and ease of purchase decision-making). Results indicate that the best-positioned websites do not necessarily correspond to the best web navigation experiences.
This research aims to analyse the current level of accuracy in predicting consumer behaviour in response to brand stimuli using artificial intelligence, comparing the results with an analysis conducted using neuromarketing biometrics. The study aims to determine the accuracy achieved in predicting consumer levels of attention and visual attraction towards visual stimuli, compared to the results recorded in a neuromarketing investigation with real users, through eye tracking. The implications of these comparative analyses are discussed in the final part of the article, concluding that the emotional intelligence tool provides very accurate predictions of consumer behaviour in response to visual stimuli. The results of this study revealed that the prediction of the percentage of users who would view each area of interest is very good, and regarding visual attraction (time until the first viewing of each area of interest), it is quite similar to the order observed by the consumer group; consequently, the level of approximation to reality of AI is very good.
