۱.
Intelligent Transportation Systems (ITS) are essential for modern urban infrastructure but grapple with real-time processing of voluminous traffic video data amid bandwidth and latency limitations. This paper introduces a novel Video Coding Machine (VCM) architecture that synergistically combines Versatile Video Coding (VVC) with an adaptive bitrate optimization algorithm—driven by neural features—and a hybrid Convolutional Neural Network (CNN)–Recurrent Neural Network (RNN) model for optimized compression and congestion prediction. The VVC core, enhanced by dynamic quantization parameter (QP) adjustments, minimizes data volume while upholding perceptual quality, whereas the CNN extracts spatial features (e.g., vehicle density) and the RNN captures temporal dynamics for precise forecasting. Evaluated on diverse real-world datasets (Cityscapes, BDD100K, Tehran traffic), the system attains 94% prediction accuracy (with 93% precision and 95% recall), 60% data reduction, and 25% faster processing versus baselines like H.264/AVC and H.265/HEVC. This framework delivers a scalable, efficient solution for smart cities, fostering real-time ITS applications, substantial cost efficiencies in storage/transmission, and improved urban mobility/safety. By bridging advanced compression and deep learning, it advances sustainable traffic management paradigms.
۲.
This research investigates the fundamental challenges inherent in traditional performance appraisal systems, such as human cognitive biases and a lack of scalability, and analyzes the application of artificial intelligence (AI) as a solution to optimize these processes. The primary objective is to present a practical framework for the responsible implementation of AI, aimed at establishing more objective, equitable, and effective appraisal systems. This study employs an integrative review methodology (searching the Scopus database from 2019 onwards) combined with qualitative thematic analysis. Based on specific inclusion criteria (i.e., a focus on socio-technical challenges), 9 specialized articles were selected for final analysis. The analysis of this corpus achieved thematic saturation. The thematic analysis led to the identification of four primary themes: (1) limitations of traditional systems; (2) key AI-driven opportunities, such as enhanced objectivity and continuous feedback; (3) critical risks (e.g., Algorithmic Bias and the Black Box Problem); and (4) implementation imperatives (e.g., the necessity of Human-in-the-Loop (HITL) Oversight and transparency). Ultimately, the study concludes that success is contingent upon human-machine synergy and proposes a three-stage Integrated Socio-Technical Systems (ISTS) Framework. This framework emphasizes Explainable AI (XAI) (XAI) and the preservation of human judgment. This study is conceptual in nature. The proposed framework offers a pathway for the sustainable and human-centric utilization of this technology, which necessitates empirical validation in future research.
۳.
The rapid growth of digital payment systems has heightened the need for accurate and scalable methods to detect credit card fraud. This study evaluates a range of machine learning and deep learning algorithms, including Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbors (KNN), XGBoost, Convolutional Neural Networks (CNN), Baseline MLP (Multi-Layer Perceptron), and Long Short-Term Memory (LSTM), to identify effective approaches for detecting fraudulent transactions. Based on comparative analysis, Random Forest and LSTM achieved the strongest individual performance, with accuracies exceeding 96%. Building on these findings, a stacking ensemble model was constructed by integrating Random Forest and LSTM as base learners and Logistic Regression as the meta-classifier. The framework incorporates Convolutional Autoencoder (CAE) for feature extraction and Random Undersampling (RUS) with three resampling ratios (1:1, 1:5, and 1:10) to address class imbalance. Experimental results show that the ensemble model provides improved predictive accuracy compared with individual algorithms, achieving an accuracy of 99.98%, precision of 99.86%, and recall of 99.89% under a 1:10 resampling ratio. Rather than proposing a new algorithmic architecture, this study contributes a systematic and unified evaluation of widely used ML and DL approaches and demonstrates the effectiveness of integrating CAE, RUS, and a Random Forest–LSTM stacking ensemble in enhancing fraud detection performance.
۴.
In the economic design of control charts, traditional approaches often assume that quality loss remains constant once the quality characteristic surpasses the specification limits. This simplification overlooks the nuanced relationship between deviation magnitude and associated costs. With the increasing adoption of Taguchi’s quality loss function in product design, which quantifies loss as a continuous function of deviation from target values, there is a compelling need to integrate this perspective into control chart methodologies. This paper addresses this gap by developing an economic design framework for control charts that incorporates variable sample sizes and sampling intervals, guided by Taguchi’s quality loss function. The objective is to optimize control chart parameters to minimize the total quality-related costs, including sampling and quality loss costs. To efficiently determine the optimal parameters, a genetic algorithm is employed, with its settings fine-tuned using Taguchi’s orthogonal arrays to enhance convergence and solution quality. The proposed model is rigorously evaluated against traditional fixed sampling interval approaches. Results demonstrate that the variable sampling strategy, informed by Taguchi’s loss function, significantly improves cost efficiency and quality control effectiveness. This integration offers a more realistic and economically sound approach to control chart design, accommodating the continuous nature of quality loss and enabling dynamic sampling adjustments. The findings underscore the potential of combining advanced optimization techniques with robust quality loss modeling to advance statistical process control practices in manufacturing and service industries.
۵.
Purpose: Indigenous sports are physical activities and games held at the local level with the participation of community members to enhance social relationships and promote individual and collective health. In the past, these sports thrived, but today they face a significant decline. The aim of this research was to investigate the factors affecting the reduction of these activities and to propose suitable solutions for their revitalization. Methodology:This study is descriptive-analytical in nature and was conducted in a field-based, practical manner. The statistical population consisted of 200 experts and specialists in the field of indigenous and community sports, from whom 127 participants were selected using stratified random sampling. Data collection was done through a questionnaire whose validity was confirmed through a review of scientific resources and consultations with experts. The reliability of the questionnaire was tested using Cronbach's alpha (α=0.83). Findings:The data were analyzed using descriptive and inferential statistical methods (including Pearson’s correlation coefficient, exploratory factor analysis, and second-order confirmatory factor analysis). The results indicated that inadequate infrastructure, cultural changes, lack of governmental support, and reduced individual and social motivation were among the key reasons for the decline of indigenous sports.Conclusion:To reconstruct and strengthen indigenous sports, it is essential to identify existing obstacles and adopt appropriate measures to create the necessary conditions for more active participation in these activities.
۶.
The growing number of exchange points in distribution systems has increased the risk of counterfeit product infiltration, posing serious threats to public health and economic stability. Existing anti-counterfeiting strategies, such as blockchain-based traceability and machine learning–driven anomaly detection, remain constrained by vulnerabilities to data manipulation and limited automation. To address these challenges, this study proposes a hybrid approach that integrates sequential pattern mining with blockchain infrastructure for trajectory-based counterfeit detection. The system applies the PrefixSpan algorithm in combination with the longest common subsequence method to detect anomalous trajectories in product distribution networks. Blockchain technology ensures immutability, transparency, and decentralized validation of distribution records, while smart contracts enable automated anomaly detection. Experimental evaluation on a real-world dataset, supplemented with simulated counterfeit trajectories, achieves an overall accuracy of 87.4% and an F1-score of 0.843, outperforming existing models. Moreover, complexity analysis demonstrates the scalability of the proposed framework by offloading computationally intensive tasks to off-chain processes.
