Publication Search

Recommender systems play a critical role in shaping user decisions across digital platforms; however, the increasing complexity of recommendation algorithms has raised serious concerns regarding transparency, trust, and accountability. This study focuses on enhancing the transparency of recommender systems by integrating Explainable Artificial Intelligence (XAI) techniques within a MovieLens-based recommendation framework. The primary problem addressed is the opacity of conventional recommendation models, which limits user understanding of why certain items are recommended and may reduce trust, perceived fairness, and system acceptance. Accordingly, the main objective of this research is to design and evaluate a hybrid explainable recommender system that balances predictive accuracy with human-understandable explanations. The proposed approach combines Matrix Factorization, feature-importance-aware neural networks, and knowledge graph embeddings to construct a robust recommendation model. To enhance explainability, multiple XAI strategies are integrated, including model-agnostic methods (LIME, SHAP, and CLIME), argumentation-based explanations, and context-aware personalized explanations. A comprehensive evaluation framework is employed, incorporating algorithmic metrics (accuracy, fidelity, robustness, counterfactual consistency, and fairness) alongside human-centered evaluations measuring trust, transparency, cognitive load, and perceived usefulness. Experimental results demonstrate that the knowledge graph–enhanced hybrid model achieves superior recommendation accuracy compared to baseline approaches. Moreover, context-aware explanations consistently outperform other methods in terms of fidelity, robustness, and user-perceived transparency, while argumentation-based explanations are found to be the most persuasive. CLIME offers a strong balance between technical stability and interpretability. The findings indicate that no single explainability technique is universally optimal; instead, hybrid and adaptive explanation strategies are most effective. In conclusion, this study confirms that human-centered, context-adaptive XAI significantly improves transparency and user trust in recommender systems, highlighting explainability as a fundamental component rather than an optional enhancement.

End-to-end autonomous driving has emerged as a promising paradigm in which deep neural networks directly map raw visual inputs to continuous control actions. Despite its effectiveness, this approach suffers from limited transparency, posing significant challenges for deployment in safety-critical driving scenarios. This study addresses the lack of interpretability in vision-based end-to-end autonomous driving systems and aims to analyze model decision-making behavior under critical conditions such as sharp steering maneuvers and abrupt control transitions. To this end, an explainable end-to-end autonomous driving framework is proposed, combining a convolutional neural network trained via imitation learning with gradient-based visual attribution techniques, including Grad-CAM. The model predicts continuous steering, throttle, and braking commands directly from front-facing camera images, while explainability mechanisms are applied to reveal input regions influencing each control decision. Model performance is evaluated using both prediction accuracy and safety-oriented behavioral metrics. Experimental results show that the proposed explainable model achieves lower control prediction errors compared to a baseline end-to-end CNN, reducing steering mean squared error from 0.034 to 0.031, throttle error from 0.021 to 0.019, and brake error from 0.018 to 0.016. Moreover, safety-oriented analysis indicates improved driving stability, with steering variance reduced from 0.087 to 0.072 and abrupt control changes decreased from 14.6 to 10.3 events. Visual explanations consistently highlight road surfaces and lane-related structures during complex maneuvers, indicating reliance on semantically meaningful cues. In conclusion, the results demonstrate that integrating explainability into end-to-end autonomous driving not only preserves predictive performance but also correlates with smoother and more stable driving behavior. This framework contributes to the development of transparent and trustworthy autonomous driving systems suitable for safety-critical applications

This study discusses the introduction of digital signature patterns using the Backpropagation method on Artificial Neural Network (JST) to identify a person's characteristics and potential. The increasing use of digital identities demands a verification system that is more secure, accurate, and adaptive to the variations of each individual's signature. The main problem faced in the signature recognition system is the low level of accuracy when the visual features of the signature have similarities between users, both in terms of shape, size, and stroke pressure. In addition, variations of signatures made by the same individual are also a challenge in the identification process. As a solution, this study implements Principal Component Analysis (PCA) to extract important features from the signature image before the training process using JST. PCA is used to reduce the data dimension so that the learning process becomes more efficient and optimal. A total of 80 signature images were used in this study, consisting of 60 training data and 20 test data. The results showed that the system was able to achieve an accuracy level of 92.5%. These findings prove that the combination of PCA and JST methods is effective in recognizing digital signature patterns and has the potential to be applied to digital security-based biometric identification systems.

This study explores the intersection of Artificial Intelligence (AI), agriculture, and decolonial philosophy, emphasizing the role of local knowledge as the foundation for sustainable agricultural technology in Indonesia. The research investigates how AI can be developed not as a tool of technological domination but as a dialogical partner that recognizes the epistemic value of indigenous wisdom. Using a mixed-method approach, the study combines algorithmic experiments applying lightweight Convolutional Neural Networks (CNN) with Explainable AI (XAI) methods such as SHAP and LIME with participatory interviews involving farmers in Bima District. Empirical findings show that models integrated with localized visualization and community-based interpretability improved user trust by 84% and reduced computational energy by 28% without compromising accuracy. More importantly, the interaction between AI and farmers revealed a form of epistemic integration where algorithmic logic aligns with traditional indicators, such as soil texture, humidity, and seasonal signs known to local farmers. Philosophically, this research asserts that sustainable AI should emerge from ecological and cultural contexts rather than imposing external frameworks. In the decolonial sense, it positions local farmers not as passive users but as active epistemic agents shaping the meaning of technology. Thus, AI becomes not only a technical instrument but a site of ethical and epistemic liberation that reaffirms human responsibility toward knowledge, culture, and the earth.

The rapid growth of open-source software (OSS) in machine learning (ML) has intensified the need for reliable, automated methods to assess project quality, particularly as OSS increasingly underpins critical applications in science, industry, and public infrastructure. This study evaluates the effectiveness of a diverse set of machine learning and deep learning (ML/DL) algorithms for classifying GitHub OSS ML projects as engineered or non-engineered using a SMOTE-enhanced and explainable modeling pipeline. The dataset used in this research includes both numerical and categorical attributes representing documentation, testing, architecture, community engagement, popularity, and repository activity. After handling missing values, standardizing numerical features, encoding categorical variables, and addressing the inherent class imbalance using the Synthetic Minority Oversampling Technique (SMOTE), seven different classifiers—K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), XGBoost (XGB), Logistic Regression (LR), Support Vector Machine (SVM), and a Deep Neural Network (DNN)—were trained and evaluated. Results show that LR (84%) and DNN (85%) outperform all other models, indicating that both linear and moderately deep non-linear architectures can effectively capture key quality indicators in OSS ML projects. Additional explainability analysis using SHAP reveals consistent feature importance across models, with documentation quality, unit testing practices, architectural clarity, and repository dynamics emerging as the strongest predictors. These findings demonstrate that automated, explainable ML/DL-based quality assessment is both feasible and effective, offering a practical pathway for improving OSS sustainability, guiding contributor decisions, and enhancing trust in ML-based systems that depend on open-source components.

Application of Multi-Layer Perceptron neural network to fault classification in high-voltage transmission lines is demonstrated in this paper. Different fault types on protected transmission line should be detected and classified rapidly and correctly. This paper presents the use of Discrete Wavelet Transform energy features combined with zero sequence current magnitude as input features for neural network classifier. The proposed method uses eight extracted features to learn hidden relationship in fault signal patterns. Using proposed approach, fault detection and classification of all 11 fault types could be achieved with high accuracy. Improved performance is experienced once the neural network is trained sufficiently with 1188 fault samples, thus performing correctly when faced with different system conditions. Results of performance studies show that proposed neural network-based classifier achieves 96.18% average accuracy, which demonstrates that it can improve the performance of conventional fault classification algorithms, which in turn can provide more efficient solutions in the management and protection of high voltage electrical systems.

Amyotrophic Lateral Sclerosis (ALS) is a progressive neurodegenerative disorder that presents significant diagnostic challenges due to its heterogeneous clinical manifestations and symptom overlap with other neurological conditions. Early and accurate diagnosis is critical for initiating timely interventions and improving patient outcomes. Traditional diagnostic approaches rely heavily on clinical expertise and manual interpretation of neuroimaging data, such as structural MRI, Diffusion Tensor Imaging (DTI), and functional MRI (fMRI), which are inherently time-consuming and prone to interobserver variability. Recent advances in Artificial Intelligence (AI) and Deep Learning (DL) have demonstrated potential for automating neuroimaging analysis, yet existing models often suffer from limited generalizability across modalities and datasets. To address these limitations, we propose a Transformer-augmented deep learning ensemble framework for automated ALS diagnosis using multi-modal neuroimaging data. The proposed architecture integrates Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Vision Transformers (ViTs) to leverage the complementary strengths of spatial, temporal, and global contextual feature representations. An adaptive weighting-based fusion mechanism dynamically integrates modality-specific outputs, enhancing the robustness and reliability of the final diagnosis. Comprehensive preprocessing steps, including intensity normalization, motion correction, and modality-specific data augmentation, are employed to ensure cross-modality consistency. Evaluation using 5-fold cross-validation on a curated multi-modal ALS neuroimaging dataset demon-strates the superior performance of the proposed model, achieving a mean classification accuracy of 94.5% ± 0.7%, precision of 93.9% ± 0.8%, recall of 92.9% ± 0.9%, F1-score of 93.4% ± 0.7%, spec-ificity of 97.4% ± 0.6%, and AUC-ROC of 0.968 ± 0.004. These results significantly outperform baseline CNN models and highlight the potential of transformer-augmented ensembles in complex neurodiagnostic applications. This framework offers a promising tool for clinicians, supporting early and precise ALS detection and enabling more personalized and effective patient management strategies.

The transition of children from early childhood education to elementary school (SD) is a critical phase in their psychological and academic development. During this phase, children face significant challenges, including changes to a more structured learning environment and increasing academic demands. At SDN 055991 in Langkat Regency, this phenomenon is reflected in the difficulties experienced by some students, particularly with basic skills such as reading, writing, and arithmetic, as well as with socializing with peers. These difficulties can impact children's long-term academic and social development. This study aims to identify the key factors influencing children's learning readiness during this transition period, utilizing artificial intelligence (AI) technology. Specifically, this study uses Artificial Neural Networks (ANN) and Decision Trees as tools to analyze the data obtained. The use of this data-driven approach allows for a more in-depth analysis of the complex patterns and relationships between various variables that influence children's learning readiness, such as family factors, social environment, and students' basic skills. This study also references various previous studies demonstrating the effectiveness of backpropagation and Deep Learning algorithms in the context of education and student performance prediction. This approach is expected to provide more precise solutions for understanding children's learning readiness and provide a more accurate picture of the factors contributing to difficulties experienced by students in the transition to elementary school. The results of this study are expected to provide relevant recommendations for parents, educators, and education policymakers to support children's learning readiness and strengthen basic education policies that are adaptive to the needs of students in this digital era.

Information technology is a technology used to manage data, including processing, acquiring, organizing, storing, and manipulating data in various ways to produce high-quality information—namely, information that is relevant, accurate, and timely. This information is used for personal, business, and governmental purposes, serving as strategic information in decision-making. To anticipate changes in weather conditions, particularly rainfall, a valid and accurate report is needed that can be useful for the public. So far, the correlation or relationship between the factors influencing weather conditions—especially rainfall—has not been precisely determined, making it mathematically difficult to create a model that can describe the correlation among all these factors. This is where Artificial Neural Networks (ANN) come into play: to create such models and map out the existing problems purely based on the input data provided. One of the capabilities of neural networks is to make predictions based on previously learned data using the backpropagation method.

Mental health is a crucial aspect of modern life, with stress and anxiety being among the most common and impactful psychological disorders. This research proposes a stress and anxiety monitoring system based on the Internet of Things (IoT), integrating biometric sensors and Deep Neural Networks (DNN) for early detection and in-depth analysis. The system is designed using MAX30102 (heart rate and SpO₂), GSR (Galvanic Skin Response), and DS18B20 (body temperature) sensors, managed by an ESP32 microcontroller and communicating through the MQTT protocol. Physiological data is collected in real-time, formatted in JSON, and transmitted to both Android and web-based applications for visualization. The DNN model is developed using the TensorFlow framework with a layered architecture and ReLU activation functions to classify four mental states: relaxed, calm, anxious, and highly stressed. The training dataset comprises both primary and secondary data, including the WESAD dataset. Model performance is evaluated through k-fold cross-validation, showing high accuracy and strong generalization capabilities. The results indicate that the integration of sensor technology and deep learning significantly improves the effectiveness of stress and anxiety detection compared to traditional methods. This system demonstrates great potential for the development of AI-based wearable devices for autonomous, real-time, and adaptive mental health monitoring.

This paper explores the application of quantum-inspired optimization algorithms in the training of large-scale Graph Neural Networks (GNNs) within distributed cloud-edge environments. GNNs have gained significant attention due to their ability to model complex relationships in graph-structured data, yet their training presents challenges such as high computational demand, inefficient resource allocation, and slow convergence, especially for large datasets. Traditional meta-heuristic algorithms, while useful, often face scalability and performance issues when applied to such large-scale tasks. To address these challenges, we propose a quantum-inspired meta-heuristic algorithm that leverages quantum principles, such as superposition and entanglement, to enhance optimization processes. The algorithm was integrated into a hybrid cloud-edge system, where computational tasks are dynamically distributed between edge nodes and the cloud, optimizing resource utilization and reducing latency. Our experimental results demonstrate significant improvements in training speed, resource efficiency, and convergence rate when compared to traditional optimization methods such as Genetic Algorithms and Simulated Annealing. The quantum-inspired algorithm not only accelerates the training process but also reduces memory usage, making it well-suited for large-scale GNN applications. Furthermore, the system's scalability was enhanced by the hybrid cloud-edge architecture, which balances computational load and enables real-time data processing. The findings suggest that quantum-inspired optimization algorithms can significantly improve the training of GNNs in distributed systems, opening new avenues for real-time applications in areas such as social network analysis, anomaly detection, and recommendation systems. Future work will focus on refining these algorithms to handle even larger datasets and more complex GNN architectures, with potential integration into edge devices for enhanced real-time decision-making.

The performance of data mining techniques has been proven accurate in many studies, but each method in data mining techniques has different accuracy depending on the type of data that is the object of research. Methods in data mining techniques are divided into several functions, namely: clustering, association, classification, and prediction, where each data mining technique objective has a superior method. Therefore, in this case the author will compare the performance of the multiple linear regression method, and neural networks with fuzzy mamdani in predicting the income of PLN Unit Takengon. In several studies, the Backpropagation method shows the highest accuracy compared to other methods. Then the prediction model with multiple linear regression also has the highest accuracy as well as the Fuzzy Mamdani method has high accuracy too. Therefore, the purpose of this study is to compare the three methods, so that it can be determined which method has a higher accuracy value. The results of this study indicate that the Back propagation method has the highest accuracy and the lowest average error, namely a MAPE value of 5.9% with an accuracy of 94.1% and an RMSE of 14398.14, followed by the multiple linear regression method obtaining a MAPE value of 6.9% with an accuracy of 93.1% and an RMSE of 15527.41, then for Fuzzy Mamdani obtaining a MAPE value of 7% with an accuracy of 93% and an RMSE of 16077.76.

The development of vocational education requires Vocational High Schools (SMK) to align their competencies with student interests and industry needs. However, a mismatch between student interests and the competencies offered can result in low enrollment, requiring schools to consider closing certain programs. This study proposes the application of Artificial Neural Networks (ANNs) as a predictive method to determine the potential closure of vocational competencies based on an analysis of student interest patterns. The data used includes interest history, academic grades, and other preference indicators, which are then subjected to a preprocessing stage to ensure the quality of the model’s input. The ANN is trained to accurately recognize interest patterns, thus generating objective and adaptive decision-making recommendations. The results show that the ANN implementation provides high accuracy in predicting student interest trends and provides more precise The development of vocational education in Vocational High Schools (SMK) requires the ability to align skill competencies with students' interests and industry needs. A mismatch between students' interests and the competencies offered can lead to low interest in certain programs, which in turn may result in the decision to close those programs. This study proposes the application of Artificial Neural Networks (ANN) as a predictive method to determine the potential closure of skill competencies based on the analysis of students' interest patterns. The data used includes interest history, academic grades, and other preference indicators. This data is processed through a preprocessing stage to ensure the quality of input for the model. The ANN is trained to accurately recognize students' interest patterns, allowing it to generate more objective and adaptive decision recommendations. The results of the study show that the application of ANN has high accuracy in predicting students' interest trends and provides more precise recommendations compared to traditional methods. Therefore, this system can be an effective tool for schools to plan curriculum policies more strategically and sustainably, as well as support decisions regarding skill programs that align with students' interests and industry needs.  

This research focuses on assessing the efficacy of a method that integrates Convolutional Neural Networks (CNN) with Decision Trees for the detection of phishing URLs. Phishing represents a major cyber threat, where cybercriminals attempt to deceive individuals into disclosing sensitive information via fraudulent websites. As the frequency of phishing attacks continues to rise, there is a pressing need for effective detection and prevention strategies. In this investigation, a dataset comprising both phishing and legitimate URLs was utilized to train a CNN-Decision Tree model. The training phase includes feature extraction from URLs using CNN, which excels at identifying intricate patterns within the data, followed by classification through Decision Trees, recognized for their capacity to deliver straightforward and comprehensible interpretations of classification outcomes. The model's performance was evaluated across nine distinct scenarios to assess its effectiveness under varying conditions. The results indicated that the hybrid CNN-Decision Tree model achieved a precision rate of 94%, a recall of 90%, and an F1-Score of 92%, with an overall accuracy of 93%. These findings suggest that the model is not only proficient in identifying phishing URLs but also maintains a commendable balance between precision and recall. This research highlights that the synergy of CNN and Decision Trees can serve as a potent solution for phishing URL detection, significantly contributing to the advancement of enhanced cybersecurity systems.

The proliferation of Internet of Things (IoT) devices has introduced significant security challenges, necessitating robust attack detection mechanisms. This study presents a comprehensive comparative analysis of ten supervised learning algorithms for IoT attack detection and classification, addressing the critical challenge of balancing detection accuracy with practical deployment constraints. Using the CICIoT2023 dataset, encompassing data from 105 IoT devices and 33 attack types, we evaluate Naive Bayes, Artificial Neural Networks (ANN), Logistic Regression (LR), k-NN, XGBoost, Random Forest (RF), LightGBM, GRU, LSTM, and CNN algorithms based on some performance metrics. The comparative test results show superior performance to the traditional ensemble approach, with RF achieving 99.29% accuracy and leading precision (82.30%), followed closely by XGBoost with 99.26% accuracy and 79.60% precision. Deep learning approaches also demonstrate strong capabilities, with CNN achieving 98.33% accuracy and 71.18% precision, though these metrics indicate ongoing challenges with class imbalance. The analysis of confusion matrices reveals varying success across different attack types, with some algorithms showing perfect detection rates for certain attacks while struggling with others. The study highlights a crucial distinction in IoT security: while high precision remains important, the potentially catastrophic impact of missed attacks necessitates equal attention to recall metrics, as evidenced by the varying recall rates across algorithms (RF: 72.19%, XGBoost: 71.69%, CNN: 64.72%). These findings provide vital insights for developing balanced, context-aware intrusion detection systems for IoT environments, emphasizing the need to consider performance metrics and practical deployment constraints.

spam is posting unsolicited messages or advertising on social media, particularly Twitter. These messages are normally designed to sell specific products and services or links. In this research, we developed a fuzzy control system to detect Arabic spam tweets based on deep learning with a large language model. Initially, we performed text cleaning and further transformed text into vectors with the help of AraGpt and AraBert. Subsequently, we employed a multi-layer perceptron network model in feature extraction of essential features. Finally, we adopted the fuzzy logic control system for classifying spam tweets using features filtered from deep networks. Employing the proposed Fuzzy logic control system provided nearly a 100% comparative to only utilizing the deep neural networks, which yielded an almost 99% throughput for both large language models Aragpt and Arabert, with a 100% F1 score for the Aragpt model and 99% for Arabert model respectively.

Advances in information and internet technologies have significantly transformed the business environment, including the financial sector. The COVID-19 pandemic has further accelerated this digital adoption, expanding the e-commerce industry and highlighting the necessity for secure online transactions. Credit Card Fraud Detection (CCFD) stands critical as the prevalence of fraudulent activities continues to rise with the increasing volume of online transactions. Traditional methods for detecting fraud, such as rule-based systems and basic machine learning models, tend to fail to keep pace with fraudsters' evolving tactics. This study proposes a novel ensemble deep learning-based approach that combines Convolutional Neural Networks (CNN), Gated Recurrent Units (GRU), and Multilayer Perceptron (MLP) with the Synthetic Minority Oversampling Technique and Edited Nearest Neighbors (SMOTE-ENN) to address class imbalance and improve detection accuracy. The methodology integrates CNN for feature extraction, GRU for sequential transaction analysis, and Multilayer Perceptron (MLP) as a meta-learner in a stacking framework. By leveraging SMOTE-ENN, the proposed approach enhances data balance and prevents overfitting. With synthetic data, the robustness and accuracy of the model have been improved, particularly in scenarios where fraudulent examples are scarce. The experiments conducted on real-world credit card transaction datasets have established that our approach outperforms existing methods, achieving higher metrics performance.

This study aims to optimize the architecture of Artificial Neural Networks (ANN) for rainfall prediction using meteorological data from Indonesia, which is known for its complex and highly variable climate patterns. Climatic variables such as temperature, humidity, air pressure, wind speed, and historical rainfall records serve as the main input features to evaluate the performance of various network configurations. Optimization is carried out by determining the appropriate number of neurons, hidden layers, activation functions, and training algorithms to enhance prediction accuracy. Model evaluation employs indicators such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) to ensure output stability. The findings indicate that a multilayer architecture combined with optimized parameters significantly improves the network’s ability to capture non-linear patterns in Indonesia’s tropical rainfall data. The optimized model produces more stable and accurate predictions compared to standard configurations. These results highlight the importance of ANN architecture optimization in supporting early warning systems, agricultural planning, water resource management, and hydrometeorological disaster mitigation across Indonesia.

This study aims to analyze the application of Deep Neural Networks (DNN) as an artificial intelligence approach in processing weather data to support more accurate and stable climate predictions. Increasingly unpredictable and fluctuating weather patterns demand modern analytical methods capable of capturing non-linear relationships among atmospheric variables. DNN is utilized due to its ability to learn complex data structures through multilayer representations that extract deeper features from input variables. Weather data such as temperature, humidity, rainfall, air pressure, and wind speed are processed through several preprocessing stages to ensure optimal model performance. This research employs a descriptive qualitative method based on literature studies to examine the role of DNN in weather prediction systems. The findings indicate that DNN demonstrates strong generalization abilities, robustness to fluctuating data, and more stable predictive outputs compared to conventional statistical approaches. Thus, DNN is considered a promising component for the development of early warning systems and modern data-driven climate analysis, offering improved reliability in understanding and forecasting atmospheric conditions.

This study aims to address the challenges and propose solutions for the Optimization of Blockchain-Based Cybersecurity Systems to Enhance Resilience Against Ransomware Attacks using a Systematic Literature Review (SLR) approach. Blockchain is increasingly recognized as a transformative technology in cybersecurity due to its decentralized structure, transparency, and robustness in securing data. Despite these advantages, its widespread adoption is hindered by several challenges, including scalability, interoperability, high energy consumption, and limited access to representative ransomware datasets. This research highlights that integrating blockchain with advanced technologies such as data analytics, machine learning, and Explainable AI (XAI) can significantly enhance its effectiveness in combating ransomware.The findings reveal that Graph Convolutional Neural Networks (GCN) enable real-time detection of ransomware patterns in network traffic with an accuracy of up to 95%. Furthermore, Layer-2 solutions like the Lightning Network and sharding effectively alleviate the load on main blockchains, thereby increasing transaction throughput. Efficient consensus mechanisms, including Proof of Stake (PoS) and Delegated Proof of Stake (DPoS), address energy consumption issues, making blockchain more adaptable to IoT and resource-constrained environments. These approaches have proven successful in enabling early detection, mitigation, and prevention of ransomware in IoT systems, cloud infrastructures, and smart grid networks. The implications of this study underscore the potential of blockchain as a critical component of proactive and adaptive cybersecurity systems. However, overcoming existing challenges requires further development of hybrid frameworks that integrate blockchain with data analytics and machine learning technologies. In addition, efforts should focus on standardizing global security protocols to enhance interoperability and creating robust, diverse ransomware datasets to support more accurate detection systems. Future research should also explore methods to minimize latency and improve blockchain efficiency in real-time cybersecurity applications.