Applied Data Science and Analysis <p style="text-align: justify;">Applied Data Science and Analysis is a respected journal dedicated to real-world applications of data science. It provides a platform for experts to share innovative ideas and methodologies. Focused on bridging theory and practice, it showcases cutting-edge research and case studies in data analysis, machine learning, and more. Welcoming diverse contributions from fields like business, healthcare, and social sciences, the journal fosters collaboration among data professionals, aiming to advance the impact of data science in practical settings</p> en-US Mon, 01 Jan 2024 00:00:00 +0000 OJS 60 Towards Trustworthy Myopia Detection: Integration Methodology of Deep Learning Approach, XAI Visualization, and User Interface System <p>Myopia, a prevalent vision disorder with potential complications if untreated, requires early and accurate detection for effective treatment. However, traditional diagnostic methods often lack trustworthiness and explainability, leading to biases and mistrust. This study presents a four-phase methodology to develop a robust myopia detection system. In the initial phase, the dataset containing training and testing images is located, preprocessed, and balanced. Subsequently, two models are deployed: a pre-trained VGG16 model renowned for image classification tasks, and a sequential CNN with convolution layers. Performance evaluation metrics such as accuracy, recall, F1-Score, sensitivity, and logloss are utilized to assess the models' effectiveness. The third phase integrates explainability, trustworthiness, and transparency through the application of Explainable Artificial Intelligence (XAI) techniques. Specifically, Local Interpretable Model-Agnostic Explanations (LIME) are employed to provide insights into the decision-making process of the deep learning model, offering explanations for the classification of images as myopic or normal. In the final phase, a user interface is implemented for the myopia detection and XAI model, bringing together the aforementioned phases. The outcomes of this study contribute to the advancement of objective and explainable diagnostic methods in the field of myopia detection. Notably, the VGG16 model achieves an impressive accuracy of 96%, highlighting its efficacy in diagnosing myopia. The LIME results provide valuable interpretations for myopia cases. The proposed methodology enhances transparency, interpretability, and trust in the myopia detection process.</p> Worood Esam Noori, A. S. Albahri Copyright (c) 2023 Applied Data Science and Analysis Thu, 23 Feb 2023 00:00:00 +0000 Fuzzy Decision-Making Framework for Sensitively Prioritizing Autism Patients with Moderate Emergency Level <p>Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disorder that requires careful assessment and management. The prioritization of ASD patients involves navigating through complexities such as conflicts, trade-offs, and the importance of different criteria. Therefore, this study focuses on prioritizing patients with ASD in the healthcare setting through an evaluation and benchmarking framework. The aim of this study is to develop a framework that utilizes Multi-Criteria Decision Making (MCDM) methods to assist healthcare professionals in prioritizing ASD patients, particularly those with moderate injury levels. The methodology of the framework outlines several phases, including dataset identification, development of a decision matrix, weighting of 19 ASD criteria using the FWZIC method, ranking 432 patients using the VIKOR method, and evaluating the proposed framework using four sensitivity analysis scenarios. Among the 19 ASD criteria, the criterion 'verbal communication' obtained the highest weight. Additionally, criteria such as 'laughing for no reason', 'nodding', 'patient movement at home', and 'pointing with the index finger' obtained similar higher weights, indicating their potential impact on ASD patients. The experimental results highlight the significance of adjusting ASD weights in influencing the final rankings obtained through the VIKOR method. This emphasizes the need for careful consideration when assigning weights to the 19 ASD criteria to ensure accurate prioritization. Moreover, the framework provides valuable insights into improving the care and support provided to individuals with autism in Iraq. The findings contribute to the existing body of knowledge in the field of autism care prioritization and pave the way for future research and interventions aimed at enhancing the quality of care for individuals with autism in Iraq.</p> Hiba Mohammed Talib, A.S. Albahri, Thierry O. C. EDOH Copyright (c) 2024 Applied Data Science and Analysis Wed, 15 Mar 2023 00:00:00 +0000 Unlocking the Potential of Autism Detection: Integrating Traditional Feature Selection and Machine Learning Techniques <p>The diagnostic process for Autism Spectrum Disorder (ASD) typically involves time-consuming assessments conducted by specialized physicians. To improve the efficiency of ASD screening, intelligent solutions based&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; on machine learning have been proposed in the literature. However, many existing ML models lack the incorporation of medical tests and demographic features, which could potentially enhance their detection capabilities by considering affected features through traditional feature selection approaches. This study aims to address the aforementioned limitation by utilizing a real dataset containing 45 features and 983 patients. To achieve this goal, a two-phase methodology is employed. The first phase involves data preparation, including handling missing data through model-based imputation, normalizing the dataset using the Min-Max method, and selecting relevant features using traditional feature selection approaches based on affected features. In the second phase, seven ML classification techniques recommended by the literature, including Decision Trees (DT), Random Forest (RF), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), AdaBoost, Gradient Boosting (GB), and Neural Network (NN), are utilized to develop ML models. These models are then trained and tested on the prepared dataset to evaluate their performance in detecting ASD. The performance of the ML models is assessed using various metrics, such as Accuracy, Recall, Precision, F1-score, AUC, Train time, and Test time. These metrics provide insights into the models' overall accuracy, sensitivity, specificity, and the trade-off between true positive and false positive rates. The results of the study highlight the effectiveness of utilizing traditional feature selection approaches based on affected features. Specifically, the GB model outperforms the other models with an accuracy of 87%, Recall of 87%, Precision of 86%, F1-score of 86%, AUC of 95%, Train time of 21.890, and Test time of 0.173. Additionally, a benchmarking analysis against five other studies reveals that the proposed methodology achieves a perfect score across three key areas. By considering affected features through traditional feature selection approaches, the developed ML models demonstrate improved performance and have the potential to enhance ASD screening and diagnosis processes.</p> Samar Hazim Hammed, A.S. Albahri Copyright (c) 2024 Applied Data Science and Analysis Mon, 01 May 2023 00:00:00 +0000 Consensus on Criteria for Selection of Sign Language Mobile Apps: A Delphi Study <p><span class="fontstyle0">In the rapidly evolving digital learning landscape, sign language mobile apps are vital in advancing sign<br>language teaching. However, ensuring the quality of these apps remains a critical challenge. To address<br>this gap, this study employs the fuzzy Delphi technique to establish a robust set of criteria for evaluating<br>the quality of sign language mobile apps. By leveraging the collective wisdom and expertise of a panel<br>of experts, the fuzzy Delphi technique facilitates a structured process for achieving consensus on the<br>essential factors contributing to evaluating sign language mobile apps. Through rigorous rounds of<br>iterative feedback and analysis, the study identifies a comprehensive list of reliable criteria<br>encompassing various dimensions, including functionality, usability, accessibility, and pedagogical<br>effectiveness. The criteria established through this method serve as a valuable resource for developers,<br>educators, and clients in selecting and developing top-notch sign language mobile apps. Developers can<br>use the criteria as a guide during the design and development stages, ensuring that their apps meet the<br>highest quality and user experience standards. Educators can rely on the criteria as a checklist for<br>evaluating and selecting appropriate apps that align with their teaching objectives and cater to the<br>diverse learning needs of their students. Clients, such as educational institutions or individuals seeking<br>sign language learning resources, can make informed decisions by referring to the established criteria,<br>promoting the adoption of clear and impactful sign language mobile apps. This study emphasizes the<br>significance of applying the fuzzy Delphi method in the context of sign language mobile app<br>assessment. Involving experts from relevant fields ensures that the established criteria capture the<br>multifaceted nature of compelling sign language learning experiences. Developing a comprehensive<br>and reliable set of criteria contributes to improving existing apps and encourages innovation in creating<br>new apps that better serve the needs of sign language learners. Overall, this research extends the<br>knowledge base of sign language teaching in the digital age by providing a robust framework for<br>assessing the quality of sign language mobile apps. The findings of this study empower stakeholders in<br>the education and technology sectors to make informed decisions, fostering the advancement of sign<br>language teaching and promoting inclusivity in digital learning environments.</span> </p> Dianese David Copyright (c) 2023 Dianese David Sat, 15 Jul 2023 00:00:00 +0000 A Quality Assessment Methodology for Sign Language Mobile Apps Using Fusion Of Enhanced Weighted Mobile App Rating Scale (MARS) and Content Expert Standardized Criteria <p>Mobile sign language apps have drawn a lot of interest recently as a way to minimize communication barriers between hearing people and people with hearing impairments. However, there are issues with the criteria and standards that should be taken into account when developing these apps. This study proposes a set of development criteria for sign language mobile apps and standardizes these criteria using the Fuzzy Delphi approach.&nbsp; Furthermore, the Fuzzy-Weighted Zero Inconsistency (FWZIC) approach is utilized to assign weights to the criteria and establish a ranking order. An initial set of requirements is developed based on the literature review. The Fuzzy Delphi technique is used, involving a panel of experts made up of developers, sign language experts, and users of sign language mobile apps, to assess the validity and reliability of the criteria. The FWZIC technique is used to give the criterion weights and determine their ranking order in order to further improve the decision-making process. The relative relevance of each criterion is determined by the FWZIC technique, which involves expert input and makes use of their knowledge and expertise. A thorough ranking is generated by taking into account the effects of each criterion on several zones, assisting in efficient decision-making during the creation of sign language mobile apps. Six Malaysian Sign Language apps that have been shortlisted are being utilized as a proof of concept to test the idea. The result of 6 apps is obtained based on the final standard criteria, their weights, and rankings.</p> <p>&nbsp;</p> Dianese David, Abdullah Hussein Copyright (c) 2023 Dianese David, Abdullah Hussein Sun, 09 Jul 2023 00:00:00 +0000 Automated Grading System for Breast Cancer Histopathological Images Using Histogram of Oriented Gradients (HOG) Algorithm <p>Breast cancer is the most common type of cancer in the world, affecting both men and women. In 2023, the American Cancer Society's reported that there will be approximately 297,800 new cases of invasive breast cancer in women and 2,850 in men, along with 55,750 cases of ductal carcinoma in situ (DCIS) in women. Further, an estimated 43,750 deaths are expected from breast cancer, of which approximately 43,180 are among women and 570 are among men. In this paper, we propose an automated grading system for breast cancer based on tumor's histopathological images using a combination of the Histogram of Oriented Gradients (HOG) for feature extraction and machine learning algorithms. The proposed system has four main phases: image preprocessing and segmentation, feature extraction, classification, and integration with a website. Grayscale conversion, enhancement, noise and artifact removal methods are used during the image preprocessing stage. Then the image is segment during the segmentation phase to extract regions of interest. And then, features are extracted from the obtained region of interest using the Histogram of Oriented Gradients (HOG) algorithm. The next, the images are classified into three distinct breast cancer grades based on the extracted features using machine learning algorithms. Moreover, the effectiveness of the proposed system was evaluated and reported using vary evaluation methods and the results showed a remarkable accuracy of up to 97% by the SVM classifier. Finally, the machine learning model is integrated into a website to improve the detection and diagnosis of breast cancer disease and facilitate the access and use of patient data. This will make the work easier for physicians to enhance breast cancer detection and treatment.</p> Mohammed Saher, Muneera Alsaedi, Ahmed Al Ibraheemi Copyright (c) 2023 Mohammed Saher, Muneera Alsaedi, Ahmed Al Ibraheemi Tue, 29 Aug 2023 00:00:00 +0000 Multi-Tiered CNN Model for Motor Imagery Analysis: Enhancing UAV Control in Smart City Infrastructure for Industry 5.0 <p>The concept of brain-controlled UAVs, pioneered by researchers at the University of Minnesota, initiated a series of investigations. These early efforts laid the foundation for more advanced prototypes of brain-controlled UAVs. However, BCI signals are inherently complex due to their nonstationary and high-dimensionality nature. Therefore, it is crucial to carefully consider both feature extraction and the classification process. This study introduces a novel approach, combining a pretrained CNN with a classical neural network classifier and STFT spectrum, into a Multi-Tiered CNN model (MTCNN). The MTCNN model is applied to decode two-class Motor Imagery (MI) signals, enabling the control of UAV up/down movement. The experimental phase of this study involved four key experiments. The first assessed the MTCNN model's performance using a substantial dataset, resulting in an impressive classification accuracy of 99.1%. The second and third experiments evaluated the model on two different datasets for the same subjects, successfully addressing challenges associated with inter-subject and intra-subject variability. The MTCNN model achieved a remarkable classification accuracy of 99.7% on both datasets. In a fourth experiment, the model was validated on an additional dataset, achieving classification accuracies of 100% and 99.6%. Remarkably, the MTCNN model surpassed the accuracy of existing literature on two BCI competition datasets. In conclusion, the MTCNN model demonstrates its potential to decode MI signals associated with left- and right-hand movements, offering promising applications in the field of brain-controlled UAVs, particularly in controlling up/down movements. Furthermore, the MTCNN model holds the potential to contribute significantly to the BCI-MI community by facilitating the integration of this model into MI-based UAV control systems.</p> Z.T. Al-Qaysi, Mahmood M. Salih, Moceheb Lazam Shuwandy, M.A. Ahmed, Yazan S.M. Altarazi Copyright (c) 2023 Z.T. Al-Qaysi, Mahmood M. Salih, Moceheb Lazam Shuwandy, M.A. Ahmed, Yazan S.M. Altarazi Wed, 20 Sep 2023 00:00:00 +0000 Developing an Electronic Health Records System Based on the National Identity by Using Angular Web Application Framework <p>Electronic Health Records systems provide a better management system for patients and the healthcare spiciest by providing an accurate, up-to-date, and complete information about patients at the point of care. Further, enabling fast access to patient records for more coordinated, efficient care. The core objective for this study is to connect the national identity with the electronic records for patients. This aims to obtain a secure and accurate identification and authentication of patients and their information as well as will enable information exchange between different healthcare centers. This research presents the design and implementation of an electronic health records (EHR) system with national ID integration to cover the mentioned significant objectives. The system utilizes the following technologies: Angular, HTML, CSS, and Bootstrap for the frontend and NestJS, Prisma, and JWT for the backend, with a Postgres database. The project follows the Agile software development methodology and includes system analysis, design, implementation, testing, and evaluation. The system features separate interfaces for administrators, doctors, and patients with appropriate table relations, endpoints, and schemas. The system was tested using Jest and test Bed for functional, performance, and security testing. The results demonstrate that the EHR system with national ID integration can improve patient record management and access. Limitations and future improvements are also discussed. Overall, this project provides a valuable contribution to the healthcare industry and offers a solid foundation for further development and refinement.</p> Mustafa Mohammed Rasheed, Muneera Alsaedi, Ahmed Al Ibraheemi Copyright (c) 2023 Mustafa Mohammed Rasheed, Muneera Alsaedi, Ahmed Al Ibraheemi Fri, 06 Oct 2023 00:00:00 +0000 The Considerations of Trustworthy AI Components in Generative AI; A Letter to Editor <p><strong>Dear Editor:</strong> In navigating the ever-expanding realms of Artificial Intelligence (AI) across diverse domains, particularly within the purviews of Generative AI, it becomes imperative to delve into the intricate considerations of trustworthy AI components [1]. This letter aims to underscore the salient aspects of this discussion with a specific focus on the thematic areas championed by the journal.</p> A. S. Albahri , Yahya Layth Khaleel, Mustafa Abdulfattah Habeeb Copyright (c) 2023 A. S. Albahri , Yahya Layth Khaleel, Mustafa Abdulfattah Habeeb Sat, 28 Oct 2023 00:00:00 +0000 Smart Real-Time IoT mHealth-based Conceptual Framework for Healthcare Services Provision during Network Failures <p>A series of healthcare problems related to frequent failures in telemedicine architecture, particularly in multi-sensors (Tier 1), medical center servers (Tier 3), and potential failures in network integration between these system components, needed to be addressed. The objective of this research was to present a novel smart real-time IoT mHealth framework within the context of IoT that could select an appropriate hospital during the aforementioned failures. The research methodology involved a new local multi-sensor fusion triage algorithm called the three-level localization triage (3LLT). This aimed to exclude the control process of patient triage and sensor fusion from the medical center, while also alarming failures related to medical body sensors. Additionally, the proposed framework was implemented using the multi-criteria decision-making (MCDM) method, connecting mHealth directly with distributed hospital servers. The distribution of hospitals was determined using the AHP (Analytic Hierarchy Process) based on the crossover between ‘healthcare services/time of arrival of the patient at the hospital’ and ‘hospitals list’ to estimate small power consumption. Validation processes were conducted for the proposed framework. The expected output from this research is to enhance the provision of healthcare services during various network failures.</p> O.S. Albahri, Amneh Alamleh , Tahsien Al-Quraishi , Rahul Thakkar Copyright (c) 2023 O.S. Albahri, Amneh Alamleh , Tahsien Al-Quraishi , Rahul Thakkar Thu, 02 Nov 2023 00:00:00 +0000 Challenges in AutoML and Declarative Studies Using Systematic Literature Review <p>Machine Learning (ML) technologies have become essential tools, transforming industries and unlocking incredible potential in various fields. ML is now widely used for data-driven decision-making and predictive analytics across fields like healthcare, finance, transportation, and more. However, building and implementing ML models can be complex and time-consuming, often requiring programming proficiency and data science skills. Despite significant progress in ML, non-experts often struggle with selecting algorithms, optimizing models, and deploying ML solutions. This paper conducts a systematic literature review to explore challenges in the area of machine learning based on multiple categories involving features engineering and data extraction, learning model structure and activities, learning-based analysis and visualization, analysis algorithms in data-based systems, machine learning algorithms and systems development, and declarative ML-based prediction. Addressing these challenges underlines the importance of following AutoML and Declarative ML strategies in simplifying the ML process.</p> Eman Thabet Khalid, Abdulla J. Y. Aldarwish, Ali A.Yassin Copyright (c) 2023 Eman Thabet Khalid, Abdulla J. Y. Aldarwish, Ali A.Yassin Wed, 15 Nov 2023 00:00:00 +0000 Application of Sequential Analysis on Runtime Behavior for Ransomware Classification <p>The unprecedented development and massive proliferation of Internet technology, computing /storage capability and emerging business model, like cloud and IoT, brings not only incredible changes to human lifestyle but also numerous, complex and continuing cyber security threats, one noticeable example among them is malware. Static analysis has been popular and widely used in many anti-virus engine. However, static analysis can be avoided using techniques such as packing, polymorphism, and metamorphism. In this paper, I propose a novel method focuses on the feature extraction, which exploits the inherent encryption behaviour of ransomwares. Specifically, runtime malicious sequential analysis is adopted to establish the desired feature set, which further facilitate the identification of the inherent encryption function. With the proposed method, an accuracy level of 96% was achieved</p> Chee Keong NG, Tahsien Al-Quraishi, Tony De Souza-Daw Copyright (c) 2023 Chee Keong NG, Tahsien Al-Quraishi, Tony De Souza-Daw Thu, 23 Nov 2023 00:00:00 +0000 A bibliometric analysis of research on multiple criteria decision making with emphasis on Energy Sector between (2019-2023) <p>In the present study, a bibliometric analysis of research works that have been conducted over the last five years in connection to Multiple Criteria Decision making (MCDM) and its application in the energy sector is presented. In the beginning, a statistical study of influential publications, journals, countries/territories, and authors was carried out. In the following step, an analysis was performed based on four distinct time periods to determine the evolving patterns of authors' cooperation structure and study themes. According to the findings, there has been a rise in the quality of collaboration between writers, as well as an increase in the number of publications and authors who have contributed to the study on MCDM during the last five years. Researchers should be able to successfully conduct investigations in linked domains with the assistance of the complete and scientific analysis of MCDM. It also concludes that there are more opportunities in the future in the field of energy applications with MCDM, and this can be encouraging for researchers from both fields, as well as those from the industrial and economic fields, to consider MCDM in their utilization of energy alternatives and to make decisions that are informed by such findings.</p> Dianese David, Abdullah Alamoodi Copyright (c) 2023 Dianese David, Abdullah Alamoodi Wed, 29 Nov 2023 00:00:00 +0000 Does Lack of Knowledge and Hardship of Information Access Signify Powerful AI? A Large Language Model Perspective <p>Large Language Models (LLMs) are evolving and expanding enormously. With the consistent improvement of LLMs, more complex and sophisticated tasks will be tackled. Handling various tasks and fulfilling different queries will be more precise. Emerging LLMs in the field of Artificial Intelligence (AI) impact online digital content. An association between digital corpus scarcity and the improvement of LLMs is drawn. The impact it will bring to the field of LLMs is discussed. More powerful LLMs are insights to be there. Specifically, increase in Reinforcement Learning from Human Feedback (RLHF) LLMs release. More precise RLHF LLMs will endure development and alternative releases.</p> Idrees A. Zahid, Shahad Sabbar Joudar Copyright (c) 2023 Idrees A. Zahid, Shahad Sabbar Joudar Tue, 12 Dec 2023 00:00:00 +0000