H.3. Artificial Intelligence
Mohammad Heydari
Abstract
Deep learning–based super-resolution has become an important tool for enhancing brain magnetic resonance imaging (MRI), particularly when acquisition constraints limit spatial resolution. Lightweight autoencoder architectures have recently been proposed to achieve computational efficiency while ...
Read More
Deep learning–based super-resolution has become an important tool for enhancing brain magnetic resonance imaging (MRI), particularly when acquisition constraints limit spatial resolution. Lightweight autoencoder architectures have recently been proposed to achieve computational efficiency while maintaining reconstruction quality. However, certain architectural choices adopted in these models—most notably the incorporation of encoder–decoder skip connections—raise methodological concerns regarding the preservation of the information bottleneck principle that defines autoencoder-based learning. This note critically examines whether such designs genuinely rely on latent representation learning or instead introduce shortcut pathways that weaken the inferential nature of super-resolution reconstruction.
H.3. Artificial Intelligence
Omid Khalaf Beigi; Seyed Alireza Bashiri Mosavi
Abstract
A brain tumor is one of the most serious and life-threatening brain diseases that can profoundly affect an individual’s life. Accordingly, the present study addresses the challenge of refining brain tumor segmentation based on Magnetic Resonance Imaging (MRI) data and deep reinforcement learning. ...
Read More
A brain tumor is one of the most serious and life-threatening brain diseases that can profoundly affect an individual’s life. Accordingly, the present study addresses the challenge of refining brain tumor segmentation based on Magnetic Resonance Imaging (MRI) data and deep reinforcement learning. Although supervised learning–based approaches have shown satisfactory performance in tumor segmentation and localization, they often suffer from high uncertainty errors along tumor boundaries. In this research, a learning framework combining a supervised model with deep reinforcement learning—referred to as DURL-Net—is proposed for segmentation and refinement purposes. Specifically, the framework first employs a U-Net architecture to generate an initial segmentation mask. This initial output and the corresponding MRI are then partitioned into localized patches, which are sequentially processed by a Deep Q-Network (DQN) agent. The DQN agent interacts with the environment by selecting optimal morphological operations (such as dilation and erosion) to refine tumor boundaries and correct uncertainties patch by patch. The dataset used in this study comprises 3,064 T1-Weighted Contrast-Enhanced MRI images, employed for both segmentation and tumor-type classification tasks. Experimental results demonstrate that DURL-Net achieved a Dice Similarity Coefficient (DSC) of 86.73%, a Jaccard Index (IoU) of 78.68%, a Kappa coefficient (Kap) of 85.21%, a Sensitivity of 87.68%, and a Specificity of 96.06%.
H.3. Artificial Intelligence
Mohammad Aalishahi; Mohammad Hadi Bokaei; Abolfazl Nadi
Abstract
Given the importance of Named Entity Recognition (NER), numerous studies have been conducted in this field. However, most research has focused on languages such as English, French, and Arabic. In contrast, studies on Persian remain limited, despite Persian being one of the most widely spoken languages ...
Read More
Given the importance of Named Entity Recognition (NER), numerous studies have been conducted in this field. However, most research has focused on languages such as English, French, and Arabic. In contrast, studies on Persian remain limited, despite Persian being one of the most widely spoken languages in West Asia, necessitating the development of NER methods for it. In this study, using Active Learning, a corpus of 1,351 advertisements from the Official Gazette was annotated. The GEMMA2b model was then fine-tuned on this data, achieving approximately 95% accuracy. This model was employed to extract around 13 types of named entities and their relationships within the advertisement texts. The primary advantage of this method is the model’s high accuracy compared to other approaches. Additionally, the use of Persian data—which, unlike languages such as English or Arabic, has fewer resources—is another notable feature of this research.
H.3. Artificial Intelligence
vahidreza afshin; Saiedeh Kabirirad; Seyed Hamid Zahiri
Abstract
Accurate and timely detection of heart disease is a fundamental challenge in clinical diagnostics, exacerbated by the inherent complexities of real-world healthcare datasets, such as class imbalance, high dimensionality and the presence of outliers. To address these issues, we propose a dual-path learning ...
Read More
Accurate and timely detection of heart disease is a fundamental challenge in clinical diagnostics, exacerbated by the inherent complexities of real-world healthcare datasets, such as class imbalance, high dimensionality and the presence of outliers. To address these issues, we propose a dual-path learning framework specifically designed for real-time heart disease detection within IoT–Fog–Cloud environments. The framework integrates advanced preprocessing techniques, including Conditional Generative Adversarial Networks (CGANs) for data balancing and a hybrid feature selection pipeline, thereby addressing data imbalance and dimensionality reduction. It also proposes a dual-path diagnostic system that uses a lightweight XGBoost classifier in the fog layer optimized for low latency and fast local inference, and an improved TabTransformer-based model in the cloud layer, which is selectively used for ambiguous samples. Experimental evaluation on heart disease datasets demonstrates the superior performance of the framework, achieving 99.66% accuracy, F1 score of 99.75%, and AUC of 99.96%. Additionally, it improves overall inference latencies, mostly below one millisecond in the fog layer and low latency in cloud layer. This paper provides a scalable, interpretable, and privacy-preserving medical diagnosis framework in distributed healthcare systems.
H.3. Artificial Intelligence
Zainab Hasan; Esmaeel Tahanian
Abstract
Today, telecommunications fraud has emerged as a major challenge for operators, resulting in billions of dollars in financial losses annually. the presence of substantial noise and severe class imbalance between legitimate and fraudulent data complicates the identification of fraud patterns within massive ...
Read More
Today, telecommunications fraud has emerged as a major challenge for operators, resulting in billions of dollars in financial losses annually. the presence of substantial noise and severe class imbalance between legitimate and fraudulent data complicates the identification of fraud patterns within massive volumes of Call Detail Records (CDRs). This paper proposes a hybrid ensemble model, termed Hybrid AdaBoost-RF, for telecommunication fraud detection. In this model, Random Forest is employed as the base learner within the AdaBoost framework to enhance the model's robustness against noise. Furthermore, the SMOTE technique is utilized to address the class imbalance problem. Additionally, we applied a decision threshold tuned on the training predictions to improve the model's sensitivity in detecting fraudulent behavior. Experimental results demonstrate that the proposed model outperforms existing methods in recent research, achieving a Recall of 0.87 and an F1-Score of 0.86 on the test partition of the evaluated CDR dataset using the adopted experimental protocol achieving a Recall of 0.87 and an F1-Score of 0.86. Moreover, the Area Under the Curve (AUC) for ROC and PR metrics reach 0.9777 and 0.8733, respectively, validating the high efficiency of the proposed model.
H.3. Artificial Intelligence
Mahdi Ahmadlou; Abolghasem Daeichian; Ali Reihanian
Abstract
The expansion of e-commerce has changed customer purchasing habits, moving them from brick-and-mortar stores to online venues. In this shift, some fundamental customer behaviors had to change because online shoppers cannot physically feel the products and rely heavily on customer reviews for evaluations. ...
Read More
The expansion of e-commerce has changed customer purchasing habits, moving them from brick-and-mortar stores to online venues. In this shift, some fundamental customer behaviors had to change because online shoppers cannot physically feel the products and rely heavily on customer reviews for evaluations. However, the lack of structured textual data poses challenges in sifting through numerous, diverse, and sometimes contradictory comments to make an informed purchasing decision. This study proposes a multi-step system for autonomously and intelligently analyzing customer comments to organize comment sections, utilizing general-purpose large language models. First, the proposed system automatically separates comments that discuss online shop services from those specifically related to products by tagging them. Then, it extracts the various product aspects discussed across all comments. Finally, these comments are categorized based on the extracted aspects. Additionally, a new labeled non-English dataset has been created as a benchmark dataset featuring tagged online-shop-related comments. The experimental results showed that the best performing model was Qwen 2.5, achieving an accuracy of 91.7 %.
H.3. Artificial Intelligence
Sara Mahmoudi Rashid
Abstract
Teleoperation systems are increasingly deployed in critical applications such as robotic surgery, industrial automation, and hazardous environment exploration. However, these systems are highly susceptible to network-induced delays, cyber-attacks, and system uncertainties, which can degrade performance ...
Read More
Teleoperation systems are increasingly deployed in critical applications such as robotic surgery, industrial automation, and hazardous environment exploration. However, these systems are highly susceptible to network-induced delays, cyber-attacks, and system uncertainties, which can degrade performance and compromise safety. This paper proposes a Graph Neural Network (GNN)-based Digital Twin (DT) framework to enhance the cyber-resilience and predictive control of teleoperation systems. The GNN-based anomaly detection mechanism accurately identifies cyber-attacks, such as false data injection (FDI) and denial-of-service (DoS) attacks, with a detection rate of 24.3% and a false alarm rate of only 1.8%, significantly outperforming conventional machine learning methods. Furthermore, the predictive digital twin model, integrated with model predictive control (MPC), effectively compensates for latency and dynamic uncertainties, reducing control errors by 14.12% compared to traditional PID controllers. Simulation results in a robotic teleoperation testbed demonstrate a 24.4% improvement in trajectory tracking accuracy under variable delay conditions, ensuring precise and stable operation.
H.3. Artificial Intelligence
Elahe Moradi
Abstract
Liver disorders are among the most common diseases worldwide, and their timely diagnosis and prediction can significantly improve treatment outcomes. In recent years, the application of artificial intelligence, particularly machine learning and deep learning algorithms, in the medical field has gained ...
Read More
Liver disorders are among the most common diseases worldwide, and their timely diagnosis and prediction can significantly improve treatment outcomes. In recent years, the application of artificial intelligence, particularly machine learning and deep learning algorithms, in the medical field has gained tremendous importance and has led to reduced healthcare costs. In this study, the ILPD dataset from the UCI Machine Learning Repository, which comprises 583 liver patient records with 11 features, was utilized. In this research, a predictive framework based on Multilayer Perceptron (MLP) is employed for the prediction of liver disorders. To address the class imbalance in the binary classification dataset, the Synthetic Minority Oversampling Technique (SMOTE)–Tomek approach was implemented to improve data balance. Moreover, due to the presence of a substantial number of outlier values, a robust scaling method was applied for their management. Finally, the performance of the proposed method was compared with three well-known machine learning algorithms. To enhance evaluation robustness, a five-fold cross-validation was employed across all classifiers. All simulations were conducted using Python, and the results illustrate that the proposed method achieves superior performance, with an accuracy of 90.90% compared to state-of-the-art approaches.
H.3. Artificial Intelligence
Soodeh Shadravan; Ali Karimi
Abstract
The Coati Optimization Algorithm (COA) is a newly developed metaheuristic algorithm, drawing inspiration from the clever tactics Coatis use when attacking Iguanas as well as their strategies for dealing with and evading predators. This algorithm has shown a commendable level of effectiveness when compared ...
Read More
The Coati Optimization Algorithm (COA) is a newly developed metaheuristic algorithm, drawing inspiration from the clever tactics Coatis use when attacking Iguanas as well as their strategies for dealing with and evading predators. This algorithm has shown a commendable level of effectiveness when compared to various other metaheuristic algorithms. Its performance metrics indicate that it outperforms many alternatives in terms of efficiency and results. To overcome challenges such as the imbalance between exploration and exploitation phases and become trapped in local optima for solving complex optimization problems, an innovative technique known as "Enhanced Opposition-Based Learning" (EOBL) has been integrated with the COA algorithm. This technique draws inspiration from Random Opposition-Based Learning methods and can effectively influence the balance between exploration and exploitation phases. The Enhanced of Coati Optimization Algorithm (EOBCOA) is a novel metaheuristic algorithm proposed to enhance the performance of the COA. This method has been applied on standard benchmark functions to improve the proposed optimization algorithm. To assess the effectiveness of the proposed EOBCOA method, it was tested on standard benchmark functions, including IEEE CEC2005, IEEE CEC2019, and seven engineering problems. The results show that the EOBCOA method outperforms other advanced algorithms in achieving global optimization.
H.3. Artificial Intelligence
Naeimeh Mohammad Karimi; Mehdi Rezaeian
Abstract
In the era of massive data, analyzing bioinformatics fields and discovering its functions are very important. The rate of sequence generation using sequence generation techniques is increasing rapidly, and researchers are faced with many unknown functions. One of the essential operations in bioinformatics ...
Read More
In the era of massive data, analyzing bioinformatics fields and discovering its functions are very important. The rate of sequence generation using sequence generation techniques is increasing rapidly, and researchers are faced with many unknown functions. One of the essential operations in bioinformatics is the classification of sequences to discover unknown proteins. There are two methods to classify sequences: the traditional method and the modern method. The conventional methods use sequence alignment, which has a high computational cost. In the contemporary method, feature extraction is used to classify proteins. In this regard, methods such as DeepFam have been presented. This research is an improvement of the DeepFam model, and the special focus is on extracting the appropriate features to differentiate the sequences of different categories. As the model improved, the features tended to be more generic. The grad-CAM method has been used to analyze the extracted features and interpret improved network layers. Then, we used the fitting vector from the transformer model to check the performance of Grad-CAM. The COG database, a massive database of protein sequences, was used to check the accuracy of the presented method. We have shown that by extracting more efficient features, the conserved regions in the sequences can be discovered more accurately, which helps to classify the proteins better. One of the critical advantages of the presented method is that by increasing the number of categories, the necessary flexibility is maintained, and the classification accuracy in three tests is higher than that of other methods.
H.3. Artificial Intelligence
Ali Nasr-Esfahani; Mehdi Bekrani; Roozbeh Rajabi
Abstract
Artificial intelligence (AI) has significantly advanced speech recognition applications. However, many existing neural network-based methods struggle with noise, reducing accuracy in real-world environments. This study addresses isolated spoken Persian digit recognition (zero to nine) under noisy conditions, ...
Read More
Artificial intelligence (AI) has significantly advanced speech recognition applications. However, many existing neural network-based methods struggle with noise, reducing accuracy in real-world environments. This study addresses isolated spoken Persian digit recognition (zero to nine) under noisy conditions, particularly for phonetically similar numbers. A hybrid model combining residual convolutional neural networks and bidirectional gated recurrent units (BiGRU) is proposed, utilizing word units instead of phoneme units for speaker-independent recognition. The FARSDIGIT1 dataset, augmented with various approaches, is processed using Mel-Frequency Cepstral Coefficients (MFCC) for feature extraction. Experimental results demonstrate the model’s effectiveness, achieving 98.53%, 96.10%, and 95.92% accuracy on training, validation, and test sets, respectively. In noisy conditions, the proposed approach improves recognition by 26.88% over phoneme unit-based LSTM models and surpasses the Mel-scale Two Dimension Root Cepstrum Coefficients (MTDRCC) feature extraction technique along with MLP model (MTDRCC+MLP) by 7.61%.
H.3. Artificial Intelligence
Vahideh Monemizadeh; Kourosh Kiani
Abstract
Anomaly detection is becoming increasingly crucial across various fields, including cybersecurity, financial risk management, and health monitoring. However, it faces significant challenges when dealing with large-scale, high-dimensional, and unlabeled datasets. This study focuses on decision tree-based ...
Read More
Anomaly detection is becoming increasingly crucial across various fields, including cybersecurity, financial risk management, and health monitoring. However, it faces significant challenges when dealing with large-scale, high-dimensional, and unlabeled datasets. This study focuses on decision tree-based methods for anomaly detection due to their scalability, interpretability, and effectiveness in managing high-dimensional data. Although Isolation Forest (iForest) and its extended variant, Extended Isolation Forest (EIF), are widely used, they exhibit limitations in identifying anomalies, particularly in handling normal data distributions and preventing the formation of ghost clusters. The Rotated Isolation Forest (RIF) was developed to address these challenges, enhancing the model's ability to discern true anomalies from normal variations by employing randomized rotations in feature space. Building on this approach, we proposed the Discrete Rotated Isolation Forest (DRIF) model, which integrates an Autoencoder for dimensionality reduction. Using a discrete probability distribution and an Autoencoder enhance computational efficiency. Experimental evaluations on synthetic and real-world datasets demonstrate that proposed model outperforms iForest, EIF, and RIF. And also achieving higher Receiver Operating Characteristic-Area Under the Curve (ROC-AUC) scores and significantly faster execution times. These findings establish the proposed model as a robust, scalable, and efficient approach for unsupervised anomaly detection in high-dimensional datasets.
H.3. Artificial Intelligence
Rasoul Hosseinzadeh; Mahdi Sadeghzadeh
Abstract
The attention mechanisms have significantly advanced the field of machine learning and deep learning across various domains, including natural language processing, computer vision, and multimodal systems. This paper presents a comprehensive survey of attention mechanisms in Transformer architectures, ...
Read More
The attention mechanisms have significantly advanced the field of machine learning and deep learning across various domains, including natural language processing, computer vision, and multimodal systems. This paper presents a comprehensive survey of attention mechanisms in Transformer architectures, emphasizing their evolution, design variants, and domain-specific applications in NLP, computer vision, and multimodal learning. We categorize attention types by their goals like efficiency, scalability, and interpretability, and provide a comparative analysis of their strengths, limitations, and suitable use cases. This survey also addresses the lack of visual intuitions, offering a clearer taxonomy and discussion of hybrid approaches, such as sparse-hierarchical combinations. In addition to foundational mechanisms, we highlight hybrid approaches, theoretical underpinnings, and practical trade-offs. The paper identifies current challenges in computation, robustness, and transparency, offering a structured classification and proposing future directions. By comparing state-of-the-art techniques, this survey aims to guide researchers in selecting and designing attention mechanisms best suited for specific AI applications, ultimately fostering the development of more efficient, interpretable, and adaptable Transformer-based models.
H.3. Artificial Intelligence
Habib Khodadadi; Vali Derhami
Abstract
The exploration-exploitation trade-off poses a significant challenge in reinforcement learning. For this reason, action selection methods such as ε-greedy and Soft-Max approaches are used instead of the greedy method. These methods use random numbers to select an action that balances exploration ...
Read More
The exploration-exploitation trade-off poses a significant challenge in reinforcement learning. For this reason, action selection methods such as ε-greedy and Soft-Max approaches are used instead of the greedy method. These methods use random numbers to select an action that balances exploration and exploitation. Chaos is commonly utilized across various scientific disciplines because of its features, including non-periodicity, unpredictability, ergodicity and pseudorandom behavior. In this paper, we employ numbers generated by different chaotic systems to select action and identify better maps in diverse states and quantities of actions. Based on our experiments on various environments such as the Multi-Armed Bandit (MAB), taxi-domain, and cliff-walking, we found that many of the chaotic methods increase the speed of learning and achieve higher rewards.
H.3. Artificial Intelligence
Thomas Njoroge Kinyanjui; Kelvin Mugoye; Rachael Kibuku
Abstract
This paper presents a Multi-Head Self-Attention Fusion Network (MHSA-FN) for real-time crop disease classification, addressing key limitations in existing models, including suboptimal feature extraction, inefficient feature recalibration, and weak multi-scale fusion. Unlike prior works that rely solely ...
Read More
This paper presents a Multi-Head Self-Attention Fusion Network (MHSA-FN) for real-time crop disease classification, addressing key limitations in existing models, including suboptimal feature extraction, inefficient feature recalibration, and weak multi-scale fusion. Unlike prior works that rely solely on CNNs or transformers, MHSA-FN integrates MobileNetV2, EfficientNetV2, and Vision Transformers (ViTs) with a structured multi-level attention framework for enhanced feature learning. A gated fusion mechanism and a Multiscale Fusion Module (MSFM) optimize local texture details and global spatial relationships. The model was trained on a combined dataset of PlantVillage and locally collected images, improving adaptability to real-world conditions. It achieved 98.66% training accuracy and 99.0% test accuracy across 76 disease classes, with 99.34% precision, 99.01% recall, and 99.04% F1 score. McNemar’s test (p = 0.125) and Bayesian superiority probability (0.851) validated its robustness. Confidence variance analysis (0.000010) outperformed existing models, demonstrating MHSA-FN as a scalable, high-performance AI solution for precision agriculture in resource-constrained environments.
H.3. Artificial Intelligence
Monireh Azimi Hemat; Ezat Valipour; Laya Ali Ahmadipoor
Abstract
Visual features extracted from images in content-based image retrieval systems are inherently ambiguous. Consequently, applying fuzzy sets for image indexing in image retrieval systems has improved efficiency. In this article, the intuitionistic fuzzy sets are used to enhance the performance of the Fuzzy ...
Read More
Visual features extracted from images in content-based image retrieval systems are inherently ambiguous. Consequently, applying fuzzy sets for image indexing in image retrieval systems has improved efficiency. In this article, the intuitionistic fuzzy sets are used to enhance the performance of the Fuzzy Content-Based Image Retrieval (F-CBIR) system. To this aim, an Intuitionistic Fuzzy Content-Based Image Retrieval (IF-CBIR) is proposed by applying intuitionistic fuzzy generators on fuzzy sets. Due to the diversity of the intuitionistic fuzzy distance measure, several are assessed in IF-CBIR; in these assessments, the measure with higher performance is identified. Finally, the proposed IF-CBIR and the existing crisp CBIR and F-CBIR simulate on Corel 5K and Corel 10K databases. The results show that our proposed method has higher (10-15%) precision compared to the mentioned methods.
H.3. Artificial Intelligence
Mohamad Mahdi Yadegar; Hossein Rahmani
Abstract
In recent years, new technologies have brought new innovations into the financial and commercial world, giving fraudsters many ways to commit fraud and cost companies big time. We can build systems that detect fraudulent patterns and prevent future incidents using advanced technologies. Machine learning ...
Read More
In recent years, new technologies have brought new innovations into the financial and commercial world, giving fraudsters many ways to commit fraud and cost companies big time. We can build systems that detect fraudulent patterns and prevent future incidents using advanced technologies. Machine learning algorithms are being used more for fraud detection in financial data. But the common challenge is the imbalance of the dataset which hinders traditional machine learning methods. Finding the best approach towards these imbalance datasets is the problem many of the researchers are facing when trying to use machine learning methods. In this paper, we propose the method called FinFD-GCN that use Graph Convolutional Networks (GCNs) for fraud detection in credit card transaction datasets. FinFD-GCN represents transactions as graph in which each node represents a transaction and each edge represents similarity between transactions. By using this graph representation FinFD-GCN can capture complex relationships and anomalies that may have been overlooked by traditional methods or were even impossible to detect with conventional approaches, thus enhancing the accuracy and robustness of fraud detection in financial data. We use common evaluation metrics and confusion matrices to evaluate the proposed method. FinFD-GCN achieves significant improvements in recall and AUC compared to traditional methods such as logistic regression, support vector machines, and random forests, making it a robust solution for credit card fraud detection. By using the GCN model for fraud detection in this credit card dataset we outperformed base models 5% and 10%, with respect to F1 and AUC, respectively.
H.3. Artificial Intelligence
Pouria Rabiei; Nosratali Ashrafi-Payaman
Abstract
Today, the amount of data with graph structure has increased dramatically. Detecting structural anomalies in the graph, such as nodes and edges whose behavior deviates from the expected behavior of the network, is important in real-world applications. Thus, in our research work, we extract the structural ...
Read More
Today, the amount of data with graph structure has increased dramatically. Detecting structural anomalies in the graph, such as nodes and edges whose behavior deviates from the expected behavior of the network, is important in real-world applications. Thus, in our research work, we extract the structural characteristics of the dynamic graph by using graph convolutional neural networks, then by using temporal neural network Like GRU, we extract the short-term temporalcharacteristics of the dynamic graph and by using the attention mechanism integrated with GRU, long-term temporal dependencies are considered. Finally, by using the neural network classifier, the abnormal edge is detected in each timestamp. Conducted experiments on the two datasets, UC Irvine messages and Digg with three baselines, including Goutlier, Netwalk and CMSketch illustrate our model outperform existing methods in a dynamic graph by 10 and 15% onaverage on the UCI and Digg datasets respectively. We also measured the model with AUC and confusion matrix for 1, 5, and 10 percent anomaly injection.
H.3. Artificial Intelligence
Fariba Taghinezhad; Mohammad Ghasemzadeh
Abstract
Artificial neural networks are among the most significant models in machine learning that use numeric inputs. This study presents a new single-layer perceptron model based on categorical inputs. In the proposed model, every quality value in the training dataset receives a trainable weight. Input data ...
Read More
Artificial neural networks are among the most significant models in machine learning that use numeric inputs. This study presents a new single-layer perceptron model based on categorical inputs. In the proposed model, every quality value in the training dataset receives a trainable weight. Input data is classified by determining the weight vector that corresponds to the categorical values in it. To evaluate the performance of the proposed algorithm, we have used 10 datasets. We have compared the performance of the proposed method to that of other machine learning models, including neural networks, support vector machines, naïve Bayes classifiers, and random forests. According to the results, the proposed model resulted in a 36% reduction in memory usage when compared to baseline models across all datasets. Moreover, it demonstrated a training speed enhancement of 54.5% for datasets that contained more than 1000 samples. The accuracy of the proposed model is also comparable to other machine learning models.
H.3. Artificial Intelligence
Seyed Alireza Bashiri Mosavi; Omid Khalaf Beigi; Arash Mahjoubifard
Abstract
Using intelligent approaches in diagnosing the COVID-19 disease based on machine learning algorithms (MLAs), as a joint work, has attracted the attention of pattern recognition and medicine experts. Before applying MLAs to the data extracted from infectious diseases, techniques such as RAT and RT-qPCR ...
Read More
Using intelligent approaches in diagnosing the COVID-19 disease based on machine learning algorithms (MLAs), as a joint work, has attracted the attention of pattern recognition and medicine experts. Before applying MLAs to the data extracted from infectious diseases, techniques such as RAT and RT-qPCR were used by data mining engineers to diagnose the contagious disease, whose weaknesses include the lack of test kits, the placement of the specialist and the patient pointed at a place and low accuracy. This study introduces a three-stage learning framework including a feature extractor by visual geometry group 16 (VGG16) model to solve the problems caused by the lack of samples, a three-channel convolution layer, and a classifier based on a three-layer neural network. The results showed that the Covid VGG16 (CoVGG16) has an accuracy of 96.37% and 100%, precision of 96.52% and 100%, and recall of 96.30% and 100% for COVID-19 prediction on the test sets of the two datasets (one type of CT-scan-based images and one type of X-ray-oriented ones gathered from Kaggle repositories).
H.3. Artificial Intelligence
Ali Rebwar Shabrandi; Ali Rajabzadeh Ghatari; Mohammad Dehghan nayeri; Nader Tavakoli; Sahar Mirzaei
Abstract
This study proposes a high-level design and configuration for an intelligent dual (hybrid and private) blockchain-based system. The configuration includes the type of network, level of decentralization, nodes, and roles, block structure information, authority control, and smart contracts and intended ...
Read More
This study proposes a high-level design and configuration for an intelligent dual (hybrid and private) blockchain-based system. The configuration includes the type of network, level of decentralization, nodes, and roles, block structure information, authority control, and smart contracts and intended to address the two main categories of challenges–operation management and data management–through three intelligent modules across the pandemic stages. In the pre-hospital stage, an intelligent infection prediction system is proposed that utilizes in-house data to address the lack of a simple, efficient, agile, and low-cost screening method for identifying potentially infected individuals promptly and preventing the overload of patients entering hospitals. In the in-hospital stage, an intelligent prediction system is proposed to predict infection severity and hospital Length of Stay (LoS) to identify high-risk patients, prioritize them for receiving care services, and facilitate better resource allocation. In the post-hospital stage, an intelligent prediction system is proposed to predict the reinfection and readmission rates, to help reduce the burden on the healthcare system and provide personalized care and follow-up for higher-risk patients. In addition, the distribution of limited Personal protective equipment (PPE) is made fair using private blockchain (BC) and smart contracts. These modules were developed using Python and utilized to evaluate the performance of state-of-the-art machine learning (ML) techniques through 10-fold cross-validation at each stage. The most critical features were plotted and analyzed using SHapely Adaptive exPlanations (SHAP). Finally, we explored the implications of our system for both research and practice and provided recommendations for future enhancements.
H.3. Artificial Intelligence
Seyed Alireza Bashiri Mosavi; Mohsen Javaherian; Omid Khalaf Beigi
Abstract
One way of analyzing COVID-19 is to exploit X-ray and computed tomography (CT) images of the patients' chests. Employing data mining techniques on chest images can provide in significant improvements in the diagnosis of COVID-19. However, in feature space learning of chest images, there exists a large ...
Read More
One way of analyzing COVID-19 is to exploit X-ray and computed tomography (CT) images of the patients' chests. Employing data mining techniques on chest images can provide in significant improvements in the diagnosis of COVID-19. However, in feature space learning of chest images, there exists a large number of features that affect COVID-19 identification performance negatively. In this work, we aim to design the dual hybrid partial-oriented feature selection scheme (DHPFSS) for selecting optimal features to achieve high-performance COVID-19 prediction. First, by applying the Zernike function to the data, moments of healthy chest images and infected ones were extracted. After Zernike moments (ZMs) segmentation, subsets of ZMs (SZMs1:n) are entered into the DHPFSS to select SZMs1:n-specific optimal ZMs (OZMs1:n). The DHPFSS consists of the filter phase and dual incremental wrapper mechanisms (IWMs), namely incremental wrapper subset selection (IWSS) and IWSS with replacement (IWSSr). Each IWM is fed by ZMs sorted by filter mechanism. The dual IWMs of DHPFSS are accompanied with the support vector machine (SVM) and twin SVM (TWSVM) classifiers equipped with radial basis function kernel as SVMIWSSTWSVM and SVMIWSSrTWSVM blocks. After selecting OZMs1:n, the efficacy of the union of OZMs1:n is evaluated based on the cross-validation technique. The obtained results manifested that the proposed framework has accuracies of 98.66%, 94.33%, and 94.82% for COVID-19 prediction on COVID-19 image data (CID) including 1CID, 2CID, and 3CID respectively, which can improve accurate diagnosis of illness in an emergency or the absence of a specialist.
H.3. Artificial Intelligence
Farid Ariai; Maryam Tayefeh Mahmoudi; Ali Moeini
Abstract
In the era of pervasive internet use and the dominance of social networks, researchers face significant challenges in Persian text mining, including the scarcity of adequate datasets in Persian and the inefficiency of existing language models. This paper specifically tackles these challenges, aiming ...
Read More
In the era of pervasive internet use and the dominance of social networks, researchers face significant challenges in Persian text mining, including the scarcity of adequate datasets in Persian and the inefficiency of existing language models. This paper specifically tackles these challenges, aiming to amplify the efficiency of language models tailored to the Persian language. Focusing on enhancing the effectiveness of sentiment analysis, our approach employs an aspect-based methodology utilizing the ParsBERT model, augmented with a relevant lexicon. The study centers on sentiment analysis of user opinions extracted from the Persian website 'Digikala.' The experimental results not only highlight the proposed method's superior semantic capabilities but also showcase its efficiency gains with an accuracy of 88.2% and an F1 score of 61.7. The importance of enhancing language models in this context lies in their pivotal role in extracting nuanced sentiments from user-generated content, ultimately advancing the field of sentiment analysis in Persian text mining by increasing efficiency and accuracy.
H.3. Artificial Intelligence
Sajjad Alizadeh Fard; Hossein Rahmani
Abstract
Fraud in financial data is a significant concern for both businesses and individuals. Credit card transactions involve numerous features, some of which may lack relevance for classifiers and could lead to overfitting. A pivotal step in the fraud detection process is feature selection, which profoundly ...
Read More
Fraud in financial data is a significant concern for both businesses and individuals. Credit card transactions involve numerous features, some of which may lack relevance for classifiers and could lead to overfitting. A pivotal step in the fraud detection process is feature selection, which profoundly impacts model accuracy and execution time. In this paper, we introduce an ensemble-based, explainable feature selection framework founded on SHAP and LIME algorithms, called "X-SHAoLIM". We applied our framework to diverse combinations of the best models from previous studies, conducting both quantitative and qualitative comparisons with other feature selection methods. The quantitative evaluation of the "X-SHAoLIM" framework across various model combinations revealed consistent accuracy improvements on average, including increases in Precision (+5.6), Recall (+1.5), F1-Score (+3.5), and AUC-PR (+6.75). Beyond enhanced accuracy, our proposed framework, leveraging explainable algorithms like SHAP and LIME, provides a deeper understanding of features' importance in model predictions, delivering effective explanations to system users.
H.3. Artificial Intelligence
Damianus Kofi Owusu; Christiana Cynthia Nyarko; Joseph Acquah; Joel Yarney
Abstract
Head and neck cancer (HNC) recurrence is ever increasing among Ghanaian men and women. Because not all machine learning classifiers are equally created, even if multiple of them suite very well for a given task, it may be very difficult to find one which performs optimally given different distributions. ...
Read More
Head and neck cancer (HNC) recurrence is ever increasing among Ghanaian men and women. Because not all machine learning classifiers are equally created, even if multiple of them suite very well for a given task, it may be very difficult to find one which performs optimally given different distributions. The stacking learns how to best combine weak classifier models to form a strong model. As a prognostic model for classifying HNSCC recurrence patterns, this study tried to identify the best stacked ensemble classifier model when the same ML classifiers for feature selection and stacked ensemble learning are used. Four stacked ensemble models; in which first one used two base classifiers: gradient boosting machine (GBM) and distributed random forest (DRF); second one used three base classifiers: GBM, DRF, and deep neural network (DNN); third one used four base classifiers: GBM, DRF, DNN, and generalized linear model (GLM); and fourth one used five base classifiers: GBM, DRF, DNN, GLM, and Naïve bayes (NB) were developed, using GBM meta-classifier in each case. The results showed that implementing stacked ensemble technique consisting of five base classifiers on gradient boosted features achieved better performance than achieved on other feature subsets, and implementing this stacked ensemble technique on gradient boosted features achieved better performance compared to other stacked ensemble techniques implemented on gradient boosted features and other feature subsets used. Learning stacked ensemble technique having five base classifiers on GBM features is clinically appropriate as a prognostic model for classifying and predicting HNSCC patients’ recurrence data.
