Yazar "Hameed, Alaa Ali" seçeneğine göre listele
Listeleniyor 1 - 20 / 34
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Abalone age prediction using machine learning(Springer Science and Business Media Deutschland GmbH, 2022) Guney, Seda; Kilinc, Irem; Hameed, Alaa Ali; Jamil, AkhtarAbalone is a marine snail found in the cold coastal regions. Age is a vital characteristic that is used to determine its worth. Currently, the only viable solution to determine the age of abalone is through very detailed steps in a laboratory. This paper exploits various machine learning models for determining its age. A comprehensive analysis of various machine learning algorithms for abalone age prediction is performed which include, backpropagation feed-forward neural network (BPFFNN), K-Nearest Neighbors (KNN), Naive Bayes, Decision Tree, Random Forest, Gauss Naive Bayes, and Support Vector Machine (SVM). In addition, five different optimizers were also tested with BPFFNN to evaluate their effect on its performance. Comprehensive experiments were performed using our data set. © 2022, Springer Nature Switzerland AG.Öğe Adaptive FEM-BPNN model for predicting underground cable temperature considering varied soil composition(Elsevier - Division Reed Elsevier India Pvt Ltd, 2024) Al-Dulaimi, Abdullah Ahmed; Guneser, Muhammet Tahir; Hameed, Alaa Ali; Marquez, Fausto Pedro Garcia; Gouda, Osama E.In underground cables of power systems, the maximum temperature of the cable is a crucial factor in determining its capacity. According to standards, the permissible operating temperature for the XLPE cable conductor under steady-state conditions is 90 degree celsius - a limit that should not be exceeded. Exceeding this temperature may lead to a thermal breakdown in the cable insulation, thereby resulting in interruption of the electrical power supply. Many factors affect the cable temperature, particularly through the processes of heat dissipation and diffusion from the cable into its surroundings. These factors include soil types and compositions, cable installation configuration, and thermo physical properties; therefore, accurate analysis of these factors is crucial for cable loading. In this study, the finite element method (FEM) is employed to predict the cable temperature considering different soil compositions and to present a new approach for the thermal analysis of an underground cable system. The novel approach considers various environmental conditions including single-layer and multi-layer soil types, homogeneous and non-homogeneous soil compositions, two configuration types - flat and trefoil - as well as two types of backfill materials, specifically sand-cement mixture backfill (SCMB) and fluidized thermal backfill (FTB), and dry zones to offer deeper insight into a thermal analysis. Given that the FEM requires the construction of a complex geometric model within an optimal operating condition to obtain results with high accuracy-a process that can often be complex as well as not adaptable because it depends on constant mathematical calculation-This paper presents a novel approach FEM-BPNN that uses an adaptive Backpropagation neural networks (BPNN) model as its mainstay. The proposed BPNN model exploits historical data from FEM to refine its predictive power, therefore, increasing its efficiency and accuracy. Furthermore, the model is subject to an optimization process, adjusting and refining its internal parameters in response to new data, with the ultimate goal of improving the predictive model capabilities for the temperature of underground power cables. The results underscored the high performance of FEM in the simulation, and it was observed that FEM yielded results closely aligned with those of the IEC standard. Moreover, the proposed FEM-BPNN demonstrated exceptional accuracy, achieving a low RMSE score of 0.008. It also exhibited impressive performance in the linear regression analysis, with an R-2 value of 0.99. These metrics collectively signify the robustness and efficacy of the model.Öğe Advanced Generative AI Methods for Academic Text Summarization(Institute of Electrical and Electronics Engineers Inc., 2024) Dar, Zaema; Raheel, Muhammad; Bokhari, Usman; Jamil, Akhtar; Alazawi, Esraa Mohammed; Hameed, Alaa AliThe exponential growth of scientific literature emphasizes the need for employing advanced techniques for effective text summarization, which can significantly speed up the research process. This study tackles the challenge by advancing scientific text summarization through AI and deep learning methods. We delve into the integration and fine-tuning of cutting-edge models, including LED-Large, Pegasus variants, and BART, aiming to refine the summarization process. Unique combinations, such as SciBERT with LED-Large, were investigated to ensure the capture of critical details frequently missed by traditional methods. This novel approach led to notable improvements in summarization effectiveness. Our findings indicate that models like LED-Large excel in quickly adapting to training data, achieving impressive semantic understanding with fewer training epochs, evidenced by achieving a FRES score of 28.5852 and ROUGE scores, including a ROUGE-l F1-Score of 0.4991. However, while extensively trained models like BART -large and Pegasus displayed strong semantic capabilities, they also pointed to the necessity for refinements in readability and higher-order n-gram overlap in the produced summaries. © 2024 IEEE.Öğe An optimized feature selection approach using sand Cat Swarm optimization for hyperspectral image classification(Elsevier, 2024) Hameed, Alaa Ali; Jamil, Akhtar; Seyyedabbasi, AmirIntegrating metaheuristic algorithms and optimization techniques with remote sensing technology has accelerated the advent of advanced methodologies for analyzing hyperspectral images (HSIs). These images, rich in detail across a broad spectral range, are pivotal for diverse applications. However, the high dimensionality of data poses challenges for obtaining optimal results therefore, a preprocessing step is necessary to reduce the dimensionality of the data to select the most effective features before the application of machine learning models. This study introduces a novel methodology that integrates Back Propagation (BP) and Variable Adaptive Momentum (BPVAM) with Sand Cat Swarm Optimization (SCSO) for the classification of hyperspectral images. Utilizing SCSO for the optimal feature selection followed by BPVAM generated more accurate classification maps. The fusion of the unique strengths of SCSO with the flexibility of BPVAM has significantly boosted the precision, efficiency, and adaptability of HSI classification. The effectiveness of our method is demonstrated using two benchmark hyperspectral datasets and validated through a comprehensive comparison with other benchmark optimization techniques, including Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO). Our findings indicate that our approach enhances classification accuracy that is comparable to the stateof-the-art methods in the domain of hyperspectral data analysis.Öğe Ant Colony and Whale Optimization Algorithms Aided by Neural Networks for Optimum Skin Lesion Diagnosis: A Thorough Review(Mdpi, 2024) Mukhlif, Yasir Adil; Ramaha, Nehad T. A.; Hameed, Alaa Ali; Salman, Mohammad; Yon, Dong Keon; Fitriyani, Norma Latif; Syafrudin, MuhammadThe adoption of deep learning (DL) and machine learning (ML) has surged in recent years because of their imperative practicalities in different disciplines. Among these feasible workabilities are the noteworthy contributions of ML and DL, especially ant colony optimization (ACO) and whale optimization algorithm (WOA) ameliorated with neural networks (NNs) to identify specific categories of skin lesion disorders (SLD) precisely, supporting even high-experienced healthcare providers (HCPs) in performing flexible medical diagnoses, since historical patient databases would not necessarily help diagnose other patient situations. Unfortunately, there is a shortage of rich investigations respecting the contributory influences of ACO and WOA in the SLD classification, owing to the recent adoption of ML and DL in the medical field. Accordingly, a comprehensive review is conducted to shed light on relevant ACO and WOA functionalities for enhanced SLD identification. It is hoped, relying on the overview findings, that clinical practitioners and low-experienced or talented HCPs could benefit in categorizing the most proper therapeutical procedures for their patients by referring to a collection of abundant practicalities of those two models in the medical context, particularly (a) time, cost, and effort savings, and (b) upgraded accuracy, reliability, and performance compared with manual medical inspection mechanisms that repeatedly fail to correctly diagnose all patients.Öğe Applications and Associated Challenges in Deployment of Software Defined Networking (SDN)(Springer Science and Business Media Deutschland GmbH, 2023) Baniya, Pashupati; Agrawal, Atul; Nand, Parma; Bhushan, Bharat; Hameed, Alaa Ali; Jamil, AkhtarSDN, a rising technology within the realm of Internet of Things (IoT), has been increasingly well-received in recent times. This article presents a summary of SDN along with its different elements, advantages, and difficulties. The paper aims to provide practical solutions for introducing OpenFlow into commercial routers without hardware modifications and extending the integration of OpenFlow with legacy control protocols and control planes. In addition, the paper presents a refactoring process for migrating traditional network applications to OpenFlow-based ones, focusing on the security challenges and techniques of open technologies like SDN, OpenROADM, and SDN-based Mobile Networks (SDMN). The document also examines the advantages and possible uses of SDMN in enhancing network adaptability, streamlining network administration, and bolstering network security. The article also discusses O-RAN network innovations and difficulties, such as AI and ML workflows that are made possible by the architecture and interfaces, security concerns, and, most importantly, standardization issues. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Öğe Artificial intelligence approach for modeling house price prediction(Institute of Electrical and Electronics Engineers Inc., 2022) Çekiç, Melihşah; Korkmaz, Kübra Nur; Mukus, Habib; Hameed, Alaa Ali; Jamil, Akhtar; Soleimani, FaezehIndexed keywords SciVal Topics Abstract Real estate has a vast market volume across the globe. This domain has been growing significantly in the past few decades. An accurate prediction can help buyers, and other decision-makers make better decisions. However, developing a model that can effectively predict house prices in complex environments is still a challenging task. This paper proposes machine learning models for the accurate prediction of real estate house prices. Furthermore, we investigated the feature importance and various data analysis methods to improve the prediction accuracy. Linear Regression, Decision Tree, XGBoost, Extra Trees, and Random Forest were used in this study. For all models, hyperparameters were first calculated using k-fold cross-validation, and then they were trained to apply to test data. The models were tested on the Boston housing dataset. The proposed method was evaluated using Root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) metrics.Öğe Assessing the spreading behavior of the Covid-19 epidemic: a case study of Turkey(Institute of Electrical and Electronics Engineers Inc., 2022) Demir, Erdem; Canıtez, Muhammed Nafiz; Elazab, Mohamed; Hameed, Alaa Ali; Jamil, Akhtar; Al-Dulaimi, Abdullah AhmedCoronavirus (Covid-19) disease is a rapidly spreading type of virus that was discovered in Wuhan, China, and emerged towards the end of 2019. During this period, various studies were conducted, and intensive studies are continued in different fields regarding coronavirus, especially in the field of medicine. The virus continues to spread and is yet to be controlled fully. Machine learning is a well-explored field in the domain of computer science that can learn patterns based on existing data and make predictions on new data. This study focused on using various machine learning approaches for predicting the spreading behavior of the COVID-19 virus. The models that were considered include SARIMAX, Extreme Gradient Boosting (XGBoost), Linear Regression (LR), Decision Tree (DT), Gradient Boosting (GB), and Artificial Neural Network (ANN). The models were trained and then predictions were made by applying these models to the daily updated data provided by the Turkish Ministry of Health. Experiments on the test data showed that both XGBoost and Decision Tree models outperformed other models.Öğe Automated Classification of Snow-Covered Solar Panel Surfaces Based on Deep Learning Approaches(Tech Science Press, 2023) Al-Dulaimi, Abdullah Ahmed; Guneser, Muhammet Tahir; Hameed, Alaa Ali; Salman, Mohammad ShukriRecently, the demand for renewable energy has increased due to its environmental and economic needs. Solar panels are the mainstay for dealing with solar energy and converting it into another form of usable energy. Solar panels work under suitable climatic conditions that allow the light photons to access the solar cells, as any blocking of sunlight on these cells causes a halt in the panels work and restricts the carry of these photons. Thus, the panels are unable to work under these conditions. A layer of snow forms on the solar panels due to snowfall in areas with low temperatures. Therefore, it causes an insulating layer on solar panels and the inability to produce electrical energy. The detection of snow-covered solar panels is crucial, as it allows us the opportunity to remove snow using some heating techniques more efficiently and restore the photovoltaics system to proper operation. This paper presents five deep learning models,-16,-19, ESNET-18, ? ESNET-50, and ? ESNET-101, which are used for the recognition and classification of solar panel images. In this paper, two different cases were applied; the first case is performed on the original dataset without trying any kind of preprocessing, and the second case is extreme climate conditions and simulated by generating motion noise. Furthermore, the dataset was replicated using the upsampling technique in order to handle the unbalancing issue. The conducted dataset is divided into three different categories, namely; all_snow, no_snow, and partial snow. The five models are trained, validated, and tested on this dataset under the same conditions 60% training, 20% validation, and testing 20% for both cases. The accuracy of the models has been compared and verified to distinguish and classify the processed dataset. The accuracy results in the first case show that the compared models-16,-19, ESNET-18, andESNET-50 give 0.9592, while R ESNET-101 gives 0.9694. In the second case, the models outperformed their counterparts in the first case by evaluating performance, where the accuracy results reached 1.00, 0.9545, 0.9888, 1.00. and 1.00 for-16,-19, R ESNET-18 and R ESNET-50, respectively. Consequently, we conclude that the second case models outperformed their peers.Öğe Auxiliary learning of non-monotonic hyperparameter scheduling system via grid search(2022) Hameed, Alaa AliRecent advancements in advanced neural networks have given rise to new adaptive learning strategies. Conventional learning strategies suffer from many issues, such as slow convergence and lack of robustness. To fully exploit its potential, these issues must be resolved. Both issues are related to the step-size, and momentum term, which is generally fixed and remains uniform for all weights associated with each network layer. In this study, the recently published Back-Propagation Algorithm with Variable Adaptive Momentum (BPVAM) algorithm has been proposed to overcome these issues and improve effectiveness for classification. The study was conducted on various hyperparameters based on the grid search approach, then the optimal values of hyperparameters have trained these algorithms. Six cases were considered with varying values of the hyperparameter to evaluate the impact of the hyperparameter on the training models. It is empirically proven that the convergence behavior of the model is improved in terms of the mean and standard deviation for accuracy and the sum of squared error (SSE). A comprehensive set of experiments indicated that the BPVAM is a robust and highly efficient algorithm.Öğe Brain Pathology Classification of MR Images Using Machine Learning Techniques(Mdpi, 2023) Ramaha, Nehad T. A.; Mahmood, Ruaa M.; Hameed, Alaa Ali; Fitriyani, Norma Latif; Alfian, Ganjar; Syafrudin, MuhammadA brain tumor is essentially a collection of aberrant tissues, so it is crucial to classify tumors of the brain using MRI before beginning therapy. Tumor segmentation and classification from brain MRI scans using machine learning techniques are widely recognized as challenging and important tasks. The potential applications of machine learning in diagnostics, preoperative planning, and postoperative evaluations are substantial. Accurate determination of the tumor's location on a brain MRI is of paramount importance. The advancement of precise machine learning classifiers and other technologies will enable doctors to detect malignancies without requiring invasive procedures on patients. Pre-processing, skull stripping, and tumor segmentation are the steps involved in detecting a brain tumor and measurement (size and form). After a certain period, CNN models get overfitted because of the large number of training images used to train them. That is why this study uses deep CNN to transfer learning. CNN-based Relu architecture and SVM with fused retrieved features via HOG and LPB are used to classify brain MRI tumors (glioma or meningioma). The method's efficacy is measured in terms of precision, recall, F-measure, and accuracy. This study showed that the accuracy of the SVM with combined LBP with HOG is 97%, and the deep CNN is 98%.Öğe Combining Text Information and Sentiment Dictionary for Sentiment Analysis on Twitter During COVID(Springer Science and Business Media Deutschland GmbH, 2024) Vidushi; Jain, Anshika; Shrivastava, Ajay Kumar; Bhushan, Bharat; Hameed, Alaa AliPresence of heterogenous huge data leads towards the ‘big data’ era. Recently, tweeter usage increased with unprecedented rate. Presence of social media like tweeter has broken the boundaries and touches the mountain in generating the unstructured data. It opened research gate with great opportunities for analyzing data and mining ‘valuable information’. Sentiment analysis is the most demanding, versatile research to know user viewpoint. Society current trend can be easily observed through social network websites. These opportunities bring challenges that leads to proliferation of tools. This research works to analyze sentiments using tweeter data using Hadoop technology. It explores the big data arduous tool called Hadoop. Further, it explains the need of Hadoop in present scenario and role of Hadoop in storing ample of data and analyzing it. Hadoop cluster, Hadoop Distributed File System (HDFS), and HIVE are also discussed in detail. The Dataset used in performing the experiment is presented. Moreover, this research explains thoroughly the implementation work and provide workflow. Next session provides the experimental results and analyzes of result. Finally, last session concludes the paper, its purpose, and how it can be used in upcoming research. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Öğe Consequential Advancements of Self-Supervised Learning (SSL) in Deep Learning Contexts(Mdpi, 2024) Abdulrazzaq, Mohammed Majid; Ramaha, Nehad T. A.; Hameed, Alaa Ali; Salman, Mohammad; Yon, Dong Keon; Fitriyani, Norma Latif; Syafrudin, MuhammadSelf-supervised learning (SSL) is a potential deep learning (DL) technique that uses massive volumes of unlabeled data to train neural networks. SSL techniques have evolved in response to the poor classification performance of conventional and even modern machine learning (ML) and DL models of enormous unlabeled data produced periodically in different disciplines. However, the literature does not fully address SSL's practicalities and workabilities necessary for industrial engineering and medicine. Accordingly, this thorough review is administered to identify these prominent possibilities for prediction, focusing on industrial and medical fields. This extensive survey, with its pivotal outcomes, could support industrial engineers and medical personnel in efficiently predicting machinery faults and patients' ailments without referring to traditional numerical models that require massive computational budgets, time, storage, and effort for data annotation. Additionally, the review's numerous addressed ideas could encourage industry and healthcare actors to take SSL principles into an agile application to achieve precise maintenance prognostics and illness diagnosis with remarkable levels of accuracy and feasibility, simulating functional human thinking and cognition without compromising prediction efficacy.Öğe CryptStego: Powerful Blend of Cryptography and Steganography for Securing Communications(Springer Science and Business Media Deutschland GmbH, 2024) Pandey, Shraiyash; Baniya, Pashupati; Nand, Parma; Hameed, Alaa Ali; Bhushan, Bharat; Jamil, AkhtarIn today’s era, security is one of the most critical issues in the development of electronic communications applications, especially when sending private data. The data may be encrypted with several algorithms; however, an extra layer of security can improve protection by a significant amount. Therefore, in this paper, we have developed an application, CryptStego, to secure data using two techniques, cryptography and steganography, to transmit data securely. The encryption of original data is executed using Blowfish algorithm, a cryptographic technique. Additionally, the encrypted data is hidden through using Least Significant Bit (LSB), a steganography technique. The implementation of both techniques offers an extensive level of security, since an intruder must firstly identify the encrypted text within the image to attain the encrypted text, then secondly to decrypt using the algorithm to obtain the original message. Therefore, any intruder must encounter multiple levels of security to obtain the original message from the cipher image. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Öğe Cyber Threat Analysis and Mitigation in Emerging Information Technology (IT) Trends(Springer Science and Business Media Deutschland GmbH, 2024) Imam, Mohsin; Wajid, Mohd Anas; Bhushan, Bharat; Hameed, Alaa Ali; Jamil, AkhtarFor the information technology sector, cybersecurity is essential. One of the main issues in the modern world is sending information from one system to another without letting the information out. Online crimes, which are on the rise daily, are the first thing that comes to mind when we think about cyber security. Various governments and businesses are adopting a number of actions to stop these cybercrimes. A lot of individuals are still quite worried about cyber security after taking many safeguards. This study’s primary goal is to examine the difficulties that modern technology-based cyber security faces, especially in light of the rising acceptance of cutting-edge innovations like server less computing, blockchain, and artificial intelligence (AI). The aim of this paper is to give readers a good overview of the most recent cyber security trends, ethics, and strategies. This study focuses on the present state of cyber security and the steps that may be taken to address the rising dangers posed by modern technology through a thorough investigation of the existing literature and actual case studies. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Öğe Deep learning for liver disease prediction(Springer Science and Business Media Deutschland GmbH, 2022) Mutlu, Ebru Nur; Devim, Ayse; Hameed, Alaa Ali; Jamil, AkhtarMining meaningful information from huge medical datasets is a key aspect of automated disease diagnosis. In recent years, liver disease has emerged as one of the commonly occurring diseases across the world. In this paper, a Convolutional Neural Network (CNN) based model is proposed for the identification of liver disease. Furthermore, the performance of CNN was also compared with traditional machine learning approaches, which include Naive Bayes (NB), Support Vector Machine (SVM), K-nearest Neighbors (KNN), and Logistic Regression (LR). For evaluation, two datasets were used: BUPA and ILPD. The experimental results showed that CNN was effective for the classification of liver disease, which produced an accuracy of 75.55%, and 72.00% on the BUPA and ILPD datasets, respectively. © 2022, Springer Nature Switzerland AG.Öğe Deep learning-based framework for monitoring of debris-covered glacier from remotely sensed images(Elsevier Science, 2022) Khan, Aftab Ahmeda; Jamil, Akhtarb; Jamil A.; Hussain, Dostdara; Ali, Imrana; Hameed, Alaa AliIn recent years, deep learning (DL) methods have proven their efficiency for various computer vision (CV) tasks such as image classification, natural language processing, and object detection. However, training a DL model is expensive in terms of both complex- ities of the network structure and the amount of labeled data needed. In addition, the imbalance among available labeled data for dif- ferent classes of interest may also adversely affect the model accuracy. This paper addresses these issues using a new convolutional neural network (CNN) based architecture. The proposed network incorporates both spatial and spectral information that combines two sub- networks: spatial-CNN and spectral-CNN. The spectral-CNN extracts spectral information, while spatial-CNN captures spatial infor- mation. Moreover, to make the features more robust, a multiscale spatial CNN architecture is introduced using different kernels. The final feature vector is formed by concatenating the outputs obtained from both spatial-CNN and spectral-CNN. To address the data imbalance problem, a generative adversarial network (GAN) was used to generate data for the underrepresented class. Finally, relatively a shallower network architecture was used to reduce the number of parameters in the network and improve the processing speed. The proposed model was trained and tested on Senitel-2 images for the classification of the debris-covered glacier. The results showed that the proposed method is well-suited for mapping and monitoring debris-covered glaciers at a large scale with high classification accuracy. In addition, we compared the proposed method with conventional machine learning approaches, support vector machine (SVM), random forest (RF) and multilayer perceptron (MLP)Öğe Deep Learning-based Semantic Search Techniques for Enhancing Product Matching in E-commerce(Institute of Electrical and Electronics Engineers Inc., 2024) Aamir, Fatima; Sherafgan, Raheimeen; Arbab, Tehreem; Jamil, Akhtar; Close Bhatti, Fazeel Nadeem; Hameed, Alaa AliSearching is the process of information retrieval utilizing specific criteria or keywords. Integrating search function-alities on e-commerce platforms enables users to efficiently locate exactly what they are searching for through keyword matching. Beyond conventional keyword matching, semantic search involves aligning products with customer queries by capturing the essence of the queries, thereby retrieving semantically related products from the pertinent catalog. Semantic search enhances the e-commerce shopping experience by allowing platforms to tailor responses to user preferences through an in-depth understanding of search intents. Challenges such as morphological variations, spelling errors, and the interpretation of synonyms, antonyms, and hypernyms are addressed through deep learning models de-signed for semantic query-product matching. This study conducts a comparative analysis of various semantic search methodologies and assesses their efficacy, incorporating deep learning strate-gies for query auto-completion and spelling corrections. The evaluation employs sentence transformer models to determine the optimal approach for semantic searching, gauged by nDCG, MRR, and MAP metrics. LSTM, BART, and n-gram models are also examined for auto-completion capabilities. The research analyzes the Amazon Shopping Queries Dataset and the Upstart Commerce catalog datasets. © 2024 IEEE.Öğe A dynamic annealing learning for PLSOM neural networks: Applications in medicine and applied sciences(Elsevier, 2023) Hameed, Alaa AliIn recent years, the field of unsupervised learning in neural networks has witnessed significant advancements. This innovative learning technique holds great promise for applications in diverse domains, with particular significance in the realms of medicine and applied sciences such as medical image analysis, drug discovery, predictive analytics, and pattern recognition. The neighborhood function plays a crucial role in the Improved Parameter-Less Self-Organizing Map (PLSOM2) algorithm by governing the rate of change in the vicinity of the winning neuron. During learning iterations, the neighborhood size is dynamically adjusted to encompass the activated neighboring neurons relative to the winning neuron. The training process begins with a larger neighborhood radius, promoting rough ordering, and gradually refines the process by reducing the radius for fine-tuning. This dynamic neighborhood size significantly influences the final training outcome of the PLSOM2 algorithm. However, one of the major bottlenecks of the PLSOM2 algorithm has been the slow ordering time and the challenge of determining an optimal neighborhood size. These issues often lead to topological defects during training, such as kinks or warps in the output maps. Merely increasing processing time has proven insufficient to overcome these challenges. In this paper, we propose a novel dynamic neighborhood function designed to accelerate the convergence process of the PLSOM2 algorithm, achieving the best shape and adaptation of the neighborhood width. The study demonstrates that by improving the neighborhood function of the PLSOM2 algorithm, map distortion can be effectively suppressed. Importantly, this enhancement enables the algorithm to handle network size, neighborhood size, and the large dimensional output space adeptly. It adaptively decreases the neighborhood size over time, ensuring convergence while appropriately managing network growth and avoiding twisting and misconfiguration. To assess the effectiveness of the proposed method, we conducted an extensive set of experiments across eight real-world benchmark datasets. Notably, the outcomes of these experiments are presented showcases the results of paired t-tests, highlighting the consistency and robustness of the proposed algorithm's performance. Despite non-significant p-values in many cases, the algorithm consistently excels across various datasets, underlining its practical significance. On the other hand, presents the results of One-way Analysis of Variance (ANOVA) tests. These results further reinforce the consistency of the performance of algorithm, as indicated by p-values exceeding the common significance threshold of 0.05. The combined findings from both tables provide strong statistical evidence of the proposed algorithm's robustness and effectiveness across diverse datasets. The proposed research introduces a dynamic neighborhood function that not only improves the PLSOM2 algorithm's convergence but also enhances its adaptability and topological preservation. These enhancements, supported by the statistical tests results, underscore the algorithm's practical significance and its suitability for real-world applications, where statistical significance may not always capture the full extent of its capabilities.Öğe Efficient artificial intelligence-based models for COVID-19 disease detection and diagnosis from CT-Scans(Institute of Electrical and Electronics Engineers Inc., 2022) Masood, Muhammad Zargham; Jamil, Akhtar; Hameed, Alaa AliCOVID-19 is contagious virus that first emerged in China in 2019's last month. It mainly infects the both the lungs and the respiratory system. The virus has severely impacted life and the economy, which exposed threats to governments worldwide to manage it. Early diagnosis of COVID-19 could help with treatment planning and disease prevention strategies. In this study, we use CT-Scanned images of the lungs to show how COVID-19 may be identified using transfer learning model and investigate which model achieved the best and fastest results. Our primary focus was to detect structural anomalies to distinguish among COVID-19 positive, negative, and normal cases with deep learning methods. Every model received training with and without transfer learning and results were compared for various versions of DenseNet and EfficientNet. Optimal results were obtained using DenseNet201 (99.75%). When transfer learning was applied, all models produced almost similar results.
