Univrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401DeepSeek vs. ChatGPT: Which Performs Better in Python Coding?12710716510.22059/jitm.2026.107165ENRania A. M.AbdallaDepartment of Information Technology, Palestine Technical University, Kadoorie, Palestine.Journal Article20260528This paper conducts a comparative evaluation of two advanced large language models (LLMs) — ChatGPT-4 and DeepSeek v3—utilizing 80 algorithmic problems from Code forces categorized into four difficulty levels: Easy (800–1100), Intermediate (1200–1600), Advanced (1700–2000), and Expert (2100–2400), focusing on code generation in Python. Standardized prompts and controlled testing conditions enable the assessment of models on accuracy, effi-ciency, and code readability. As the complexity of issues increases, DeepSeek frequently out-performs ChatGPT in both accuracy and efficiency, despite both models excelling in simpler tasks. This, however, results in reduced code clarity and increased memory use. While less pre-cise at elevated levels, ChatGPT produces more concise and idiomatic responses. Both models had limited competence at the expert level; however, DeepSeek-R1 indicated a slight edge. The study illustrates a trade-off between accuracy and code clarity, so as to inform the selection of LLMs based on task requirements and provide a foundation for future efforts in optimizing code generation models for actual applications.https://jitm.ut.ac.ir/article_107165_be406ea61e97982d5919f1a6d76e529f.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401An Ensemble Machine Learning Approach for Pre-IVF Prediction of Live Birth Outcomes284310716610.22059/jitm.2026.107166ENPriyaM KResearch scholar, School of Computer Science and Engineering, REVA University, Bengaluru, Karntaka, IndiaLithinKumbleAssociate Professor, School of Computer Science and Engineering, REVA University, Bengaluru, Karntaka, India.Journal Article20260528This study aims to develop an AI driven model to predict in vitro fertilization (IVF) outcomes and improve the cost effectiveness of reproductive treatments. The primary objective is to estimate the likelihood of a live birth from the outset using advanced data processing techniques. Prediction models were applied to forecast live births among women undergoing their first cycle of fresh or frozen IVF or intracytoplasmic sperm injection (ICSI), incorporating both pre cycle and post cycle factors. A key focus of this work is predicting live birth probability when embryos originate from a couple rather than a donor. Using the publicly accessible Human Fertilization and Embryology Authority (HFEA) dataset, we evaluated several AI approaches, including Random Forest, Gradient Boosting, and a proposed ensemble machine learning algorithm. Data were preprocessed using the Auto Label Encoder technique. Model performance was assessed through confusion matrices, F1 scores, precision, recall, and receiver operating characteristic (ROC) curves. The ensemble algorithm achieved the strongest overall performance, with accuracy of 79, precision of 77, recall of 76, and an F1 score of 76.49. Gradient Boosting demonstrated the highest recall (80.48) but showed average performance in other metrics. Random Forest yielded comparatively lower accuracy and F1 scores, indicating limitations in class differentiation. These findings suggest that ensemble based AI models offer a more robust approach for predicting live birth outcomes in IVF treatment pathways.https://jitm.ut.ac.ir/article_107166_48fddf196b2205f8b335c0a295bdbf74.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401Challenges and Success Stories in Technology Adoption445810716710.22059/jitm.2026.107167ENSyeda KauserFatimaDepartment of General Education, College of Liwa, Abu Dhabi University, Abu Dhabi, UAE.FaizaQasmiDepartment of General Education, College of Liwa, Abu Dhabi University, Abu Dhabi, UAE.Journal Article20260528The rapid adoption of technology in higher education has had a profound impact on instructional strategies, learning experiences, and administrative procedures. This study examines the challenges and achievements of technology integration in higher education, focusing on organizational, cultural, technological, and skill-related barriers. Comparative studies indicate that although resistance to change, insufficient funding, and inadequate infrastructure present significant challenges, these bar-riers can be transformed into opportunities for development and innovation through proactive measures, institutional support, and strategic leadership. The potential of digital technology to im-prove academic achievement, support flexible learning models, and increase student engagement is demonstrated through case studies of successful implementations, such as Learning Management Systems (LMS), virtual classrooms, and gamification applications. Leadership commitment, contin-uous professional development, student involvement, and regular monitoring and evaluation have been highlighted as critical success factors. The paper also discusses the implications of future tech-nology adoption, including the growing influence of artificial intelligence, hybrid learning models, and the need for student-centered educational approaches. Higher education institutions can build resilient ecosystems that foster innovation and continuous improvement by utilizing data-driven de-cision-making, strategic planning, and stakeholder engagement. This study seeks to advance under-standing of the complex landscape of technology adoption in higher education and provide valuable insights for educators, administrators, and policymakers.https://jitm.ut.ac.ir/article_107167_97101bb266144f1bf56810f6c1b12a71.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401Stochastic Decision-Support Modeling for Digital Supply Chain Management under Demand and Lead-Time Uncertainty597610716810.22059/jitm.2026.107168ENMarynaIurchenkoAssociate Prof., Department of Informatics and Statistics, Klaipeda University, Klaipeda, 92294, H. Manto, 84, Lithuania.VitalinaBabenkoProf., Kharkiv National University of Radio Electronics, Kharkiv 61166, Ukraine; Kharkiv National Automobile and Highway University, Kharkiv, 61002, Ukraine.LiudmylaLomovskykhProf., Department of Global Economics, State Biotechnological University, Kharkiv, Ukraine.TetianaKlymenkoAssociate Prof., Department of Business, Administration and Law, Higher Educational Institution «University of Future Transformation», Chernihiv, Ukraine.Journal Article20260528In digital supply chain management, the effectiveness of managerial decision-making depends on the ability of information systems to support timely and cost-efficient delivery planning under uncertainty. This study develops a stochastic decision-support model for product delivery planning under random consumer demand and variable delivery lead time. The proposed approach addresses the limitations of deterministic inventory models, which often fail to reflect the uncertainty of real logistics in information-management environments. The model minimizes expected total costs by balancing inventory holding costs and losses caused by stockouts. Deviations between scheduled and actual delivery time, as well as between expected and actual inventory depletion time, are represented as continuous normally distributed random variables. This enables the analytical derivation of the expected cost function and reduces the optimization problem to determining the optimal scheduled delivery time. The optimality condition is obtained through an integral equation and is shown to have a unique solution due to the monotonic behavior of the corresponding probability function. A numerical example and graphical analysis demonstrate how holding costs, shortage losses, and uncertainty levels affect delivery timing decisions. The results show that higher holding costs shift the optimal delivery time toward later deliveries, whereas higher shortage losses require earlier scheduling. Increased uncertainty strengthens the sensitivity of the decision-support model to planning errors. The proposed model can be used as an analytical component of digital inventory-control and supply chain management systems.https://jitm.ut.ac.ir/article_107168_b3c1ccb9165d152a7685c8b52b3f2fed.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401Web Accessibility Compliance of Saudi Higher Education Institutions: A Comparative Automated Evaluation using A Checker, TAW, and WAVE779910716910.22059/jitm.2026.107169ENNadiaAl-GhreimilCollege of Computer and Information Sciences, King Saud University, Riyadh 11451, Saudi Arabia.DuaaAlSaeedCollege of Computer and Information Sciences, King Saud University, Riyadh 11451, Saudi Arabia.Heyam H.Al-BaityCollege of Computer and Information Sciences, King Saud University, Riyadh 11451, Saudi Arabia.FatimahAlmohaimeedCollege of Computer and Information Sciences, King Saud University, Riyadh 11451, Saudi Arabia.Journal Article20260528This study aims to evaluate the compliance of 44 Saudi university websites with the Web Content Accessibility Guidelines (WCAG). The analysis was conducted by evaluating the homepages of 44 Saudi university websites against the Web Content Accessibility Guidelines (WCAG). The evaluation was based on the four core accessibility principles—Perceivable, Operable, Understandable, and Ro-bust (POUR)—and assessed across three conformance levels: A, AA, and AAA. The adopted ap-proach was to use three automated heuristic evaluation tools, namely, A Checker, TAW, and WAVE. The results of the study showed that all evaluated websites failed to comply with the WCAG guide-lines, with major violations found at level A in the Perceivable and Operable principles. It also showed wide variability in issues detected by the three tools. A large number of issues are related to missing alternative text for images or fields, as well as color contrast errors. The findings suggest that Saudi university websites are not accessible to individuals with disabilities, which limits their ability to access educational content and services. A substantial proportion of the identified accessi-bility issues could be addressed by providing appropriate alternative text and labels, as well as im-proving color contrast. Additional improvements include ensuring the proper semantic use of empha-sis and strong tags instead of relying solely on italic and bold formatting. Furthermore, using multi-ple tools in the evaluation can highlight different issues, but additional expert opinions are needed. This study can be beneficial for future improvements in the accessibility of Saudi university web-sites and contribute to a more inclusive web environment.https://jitm.ut.ac.ir/article_107169_c4f13ec437e5155fcb290653a498e865.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401AI-Driven Transformation in Libraries: AI Features in Open-Source vs. Proprietary Applications10010910723210.22059/jitm.2026.107232ENKadiriRaihanathSenior Librarian, Liwa College, Abu Dhabi, UAE.AnjaliMithu GayanAssistant Professor, Dept. of Library and Information Science, Tripura University, Suryamaninagar, Tripura, India.Journal Article20260601Libraries are fast-growing with the implementation of Artificial Intelligence technologies, trans-forming operations, resource management, and user services. This study examines AI implementa-tion in open-source and proprietary library systems. AI-driven solutions enhance cataloguing, metadata management, search functionality, discovery services, virtual assistance, and personalized user experiences. Proprietary systems are competitive in exploring AI adoption with predictive ana-lytics, automated classification, linked data, discovery services, and virtual assistants. In contrast, open-source systems show potential but require further technological advancements in the current studies. The study also explores the role of AI in digital repositories, research management, and dis-covery services. Future developments should try to improve open-source AI tools to ensure wider accessibility and inclusion of various types of libraries.https://jitm.ut.ac.ir/article_107232_c67c28461985d3c11b3a0468378533d0.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401Blockchain-Based Smart Multimodal Biometric Multimedia Transmission11011910723310.22059/jitm.2026.107233ENGhadaAlhudhudProfessor, Department of Data Science and Artificial Intelligence, Faculty of Information Technology, Al-Ahilya Amman University, Amman, Jordan.MarwanAl-AkaidiProfessor, Vice Chancellor and CEO Wigwe University, Isiokpo, NigeriaJournal Article20260601In recent years, the rapid growth of freely accessible digital and multimedia content, coupled with declining trust in online security, has created an urgent need for more reliable and efficient transmis-sion frameworks. Conventional protection methods—such as passwords and periodically updated encryption keys—have proven inadequate against escalating cyber threats, identity spoofing, and da-ta breaches. This study presents a novel framework that integrates blockchain technology with seam-less biometric authentication to ensure secure data transmission. The system employs multimodal biometric and facial recognition techniques to encrypt multimedia content, while a blockchain ledger verifies, traces, and records all data exchanges within a self-sufficient validation environment. Through architectural design, simulation, and threat modeling, results show that combining multi-modal biometric fusion with blockchain significantly enhances authentication reliability and data integrity, minimizing manipulation and unauthorized access. Unlike traditional systems, the pro-posed framework strengthens data consistency, user identity assurance, and protection from cyberat-tacks—delivering a superior user experience well-suited for next-generation multimedia applica-tions.https://jitm.ut.ac.ir/article_107233_4ec91f1bb0d1f928513f5fdc074df4c8.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401Advanced Information Retrieval Techniques in the Big Data Era: Trends, Challenges, and Applications12014810723410.22059/jitm.2026.107234ENAbdulaziz Bin FahadBin Mogren AlsaudDepartment of Information Science, Faculty of Arts and Humanities, King Abdulaziz University, Jeddah, Saudi Arabia.EzzatMansourDepartment of Information Science, Faculty of Arts and Humanities, King Abdulaziz University, Jeddah, Saudi Arabia.Journal Article20260601The rapid expansion of Big Data has introduced novel opportunities and challenges for Information Retrieval (IR). This study examines the current state of IR techniques and their evolution to manage, organize, and derive meaningful insights from massive datasets. We explore how machine learning algorithms, deep learning models, and natural language processing (NLP) enhance data retrieval ac-curacy and velocity. A comprehensive analysis of contemporary methodologies indicates that per-sonalized search engines, e-commerce, and healthcare offer significant potential for improving re-trieval precision, scalability, and relevance. Furthermore, this study addresses critical ethical consid-erations, including data privacy and algorithmic bias, while exploring novel applications in autono-mous systems and personalized AI assistants. Advancing IR methodologies is vital in the Big Data era. Future research must focus on developing novel algorithmic procedures, integrating quantum computing, and establishing ethical AI practices. Ultimately, accelerating IR advancements is essen-tial to overcoming Big Data constraints and fostering technological innovation.https://jitm.ut.ac.ir/article_107234_81f8ed564710920e3c8760fd1dc1d36e.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401The Impact of Information and Communications Technology on Entrepreneurial Orientation in North Cyprus14916610723510.22059/jitm.2026.107235ENMusbahJ. AqelDepartment of Computer Science, Faculty of Information Technology, Zarqa University, Jordan.WalidEmarDepartment of Renewable Energy, Faculty of Engineering Technology, Zarqa University, Jordan.AhmadAzzaziDepartment of Extended-Reality and Games Development, Faculty of Information Technology, Applied Science Private University, Amman, Jordan.AkinsolaDanielDepartment of Management Information Systems School of Information Systems Cyprus International University Lefkosa, North Cyprus.Journal Article20260601This study examines the influence of information and communication technology (ICT) on the di-mensions of entrepreneurial orientation (EO), specifically innovativeness, risk-taking, and proac-tiveness. The investigation was motivated by evidence from developed economies indicating a pos-itive association between ICT adoption and entrepreneurial orientation. Drawing on Cartesian and contingency theory perspectives, three regression-based models were formulated to assess the pro-posed relationships. Data were collected through structured questionnaires distributed both online and in person to international student entrepreneurs. A total of 146 valid responses were obtained and analyzed using simple regression techniques. The findings reveal that ICT has a positive, statis-tically significant effect on all three EO dimensions: risk-taking, proactiveness, and innovativeness. By simultaneously testing three distinct models, this study offers a structured analytical approach that has not previously been applied to these relationships. The empirical results provide a founda-tion for recommendations regarding the strategic role of ICT in fostering entrepreneurial orienta-tion.https://jitm.ut.ac.ir/article_107235_13585afe490d7d5af1a54718dd323a10.pdfUnivrsity Of Tehran PressJournal of Information Technology Management2980-797218220260401Secure Multi-Sensor IoT-Based Fire Detection System Using MQTT and TLS Encryption16717510723610.22059/jitm.2026.107236ENSaraSeyamStudent, Department of General Education, Liwa University, Abu Dhabi, UAE.SimaSabouniStudent, Department of General Education, Liwa University, Abu Dhabi, UAE.RashaHassanAssistant Prof., Senior Lecturer, Department of General Education, Liwa University, Abu Dhabi, UAE.Journal Article20260601This paper presents an IoT-based fire detection system that integrates multiple sensors (MQ2, DHT11, and PIR) with an MQTT communication protocol to enhance detection accuracy and reduce false alarms. By combining multiple sensors, manual override mechanisms, and a real-time dash-board interface, the system allows emergency response while providing flexibility for user interven-tion in case of failures. The system is protected with TLS encryption in order to protect the data in-tegrity in such a crucial system. The tests taken regarding this system confirm the reliability of the alerts and the robustness of the system performance in simulated scenarios.https://jitm.ut.ac.ir/article_107236_2bee746a72191357760cb54855b2ee5a.pdf