Abdulaziz Alhamadani

Abdulaziz Alhamadani

Auto-generated profile, based on publicly available information

Abdulaziz Alhamadani

Florida Polytechnic University

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 March 2024 in Heliyon
Verified
Journal Article
Open access
Walaa Magdy AhmedWalaa Magdy Ahmed
Walaa Magdy Ahmed
AAAmr Ahmed Azhari
Amr Ahmed Azhari
AAAmal Alfaraj
Amal Alfaraj
AAAbdulaziz Alhamadani
Abdulaziz Alhamadani
MZMin Zhang
Min Zhang
CLChang-Tien Lu
Chang-Tien Lu
AI technology presents a variety of benefits and challenges for educators. To investigate whether ChatGPT and Google Bard (now is named Gemini) are valuable resources for generating multiple-choice questions for educators of dental caries. A book on dental caries was used. Sixteen paragraphs were extracted by an expert consultant based on applicability and potential for developing multiple-choice questions. ChatGPT and Bard language models were used to produce multiple-choice questions based on this input, and 64 questions were generated. Three dental specialists assessed the relevance, accuracy, and complexity of the generated questions. The questions were qualitatively evaluated using cognitive learning objectives and item writing flaws. Paired sample t-tests and two-way analysis of variance (ANOVA) were used to compare the generated multiple-choice questions and answers between ChatGPT and Bard. There were no significant differences between the questions generated by ChatGPT and Bard. Moreover, the analysis of variance found no significant differences in question quality. Bard-generated questions tended to have higher cognitive levels than those of ChatGPT. Format error was predominant in ChatGPT-generated questions. Finally, Bard exhibited more absolute terms than ChatGPT. ChatGPT and Bard could generate questions related to dental caries, mainly at the cognitive level of knowledge and comprehension. Language models are crucial for generating subject-specific questions used in quizzes, tests, and education. By using these models, educators can save time and focus on lesson preparation and student engagement instead of solely focusing on assessment creation. Additionally, language models are adept at generating numerous questions, making them particularly valuable for large-scale exams. However, educators must carefully review and adapt the questions to ensure they align with their learning goals.

18 July 2023
Verified
Conference Paper
Open access
JHJianfeng He
Jianfeng He
SWSyuan-Ying Wu
Syuan-Ying Wu
AAAbdulaziz Alhamadani
Abdulaziz Alhamadani
CCChih-Fang Chen
Chih-Fang Chen
WLWen-Fang Lu
Wen-Fang Lu
YLYanlin Li
Yanlin Li
Metro systems are vital to our daily lives, but they face safety or reliability challenges, such as criminal activities or infrastructure disruptions, respectively. Real-time threat detection and analysis are crucial to ensure their safety and reliability. Although many existing systems use Twitter to detect metro-related threats or events in real-time, they have limitations in event analysis and system maintenance. Specifically, they cannot analyze event development, or prioritize events from numerous tweets. Besides, their users are required to continuously monitor system notifications, use inefficient content retrieval methods, and perform detailed system maintenance. We addressed those issues by developing the MetroScope system, a real-time threat/event detection system applied to Washington D.C. metro system. MetroScope can automatically analyze event development, prioritize events based on urgency, send emergency notifications via emails, provide efficient content retrieval, and self-maintain the system. Our MetroScope system is now available at http://orion.nvc.cs.vt.edu:5000/, with a video (https://www.youtube.com/watch?v=vKIK9M60-J8) introducing its features and instructions. MetroScope is a significant advancement in enhancing the safety and reliability of metro systems.

4 August 2023
Verified
Conference Paper
Open access
JHJianfeng He
Jianfeng He
XZXuchao Zhang
Xuchao Zhang
SLShuo Lei
Shuo Lei
AAAbdulaziz Alhamadani
Abdulaziz Alhamadani
FCFanglan Chen
Fanglan Chen
CLChang-Tien Lu
Chang-Tien Lu
Few-shot text classification has extensive application where the sample collection is expensive or complicated. When the penalty for classification errors is high, such as early threat event detection with scarce data, we expect to know "whether we should trust the classification results or reexamine them.'' This paper investigates the Uncertainty Estimation for Few-shot Text Classification (UEFTC), an unexplored research area. Given limited samples, a UEFTC model predicts an uncertainty score for a classification result, which is the likelihood that the classification result is false. However, many traditional uncertainty estimation models in text classification are unsuitable for implementing a UEFTC model. These models require numerous training samples, whereas the few-shot setting in UEFTC only provides a few or just one support sample for each class in an episode. We propose Contrastive Learning from Uncertainty Relations (CLUR) to address UEFTC. CLUR can be trained with only one support sample for each class with the help of pseudo uncertainty scores. Unlike previous works that manually set the pseudo uncertainty scores, CLUR self-adaptively learns them using our proposed uncertainty relations. Specifically, we explore four model structures in CLUR to investigate the performance of three common-used contrastive learning components in UEFTC and find that two of the components are effective. Experiment results prove that CLUR outperforms six baselines on four datasets, including an improvement of 4.52% AUPR on an RCV1 dataset in a 5-way 1-shot setting. Our code and data split for UEFTC are in https://github.com/he159ok/CLUR_UncertaintyEst_FewShot_TextCls.

27 January 2026 in Dentistry Journal
Verified
Journal Article
Open access
PDF file
AAAmr Ahmed Azhari
Amr Ahmed Azhari
Walaa Magdy AhmedWalaa Magdy Ahmed
Walaa Magdy Ahmed
AAAbdulaziz Alhamadani
Abdulaziz Alhamadani
AAAmal Alfaraj
Amal Alfaraj
MZMin Zhang
Min Zhang
CLChang-Tien Lu
Chang-Tien Lu
Objective: This study aimed to compare the accuracy of two large language models (LLMs)—ChatGPT (version 3.5) and Google Gemini (formerly Bard)—in answering dental caries-related multiple-choice questions (MCQs) using a simulated student examination framework across seven examination lengths. Materials and Methods: A total of 125 validated dental caries MCQs were extracted from Dental Decks and Oxford University Press question banks. Seven examination groups were constructed with varying question counts (25, 35, 45, 55, 65, 75, and 85 questions). For each group, 100 simulations were generated per LLM (ChatGPT and Gemini), resulting in 1400 simulated examinations. Each simulated student received a unique randomized subset of questions. MCQs were answered by each LLM using a standardized prompt to minimize ambiguity. Outcomes included mean score, passing rate (≥60%), and performance differences between LLMs. Statistical analyses included independent t-tests, one-way ANOVA within each LLM, and two-way ANOVA examining interactions between LLM type and question count. Results: Across all seven examination formats, Gemini significantly outperformed ChatGPT (p < 0.001). Gemini achieved higher passing rates and higher mean scores in every examination length. One-way ANOVA revealed significant score variation with increasing exam length for both LLMs (p < 0.05). Two-way ANOVA demonstrated significant main effects of LLM type and question count, with no significant interaction. Randomization had no measurable effect on Gemini performance but influenced ChatGPT scores. Conclusions: Gemini demonstrated superior accuracy and higher passing rates compared to ChatGPT in all simulated examination formats. While both LLMs struggled with complex caries-related content, Gemini provided more reliable performance across question quantities. Educators should exercise caution in relying on LLMs for automated assessment or self-study, and future research should evaluate human–AI hybrid models and LLM performance across broader dental domains.

17 June 2025
Verified
Conference Paper
Open access
PDF file
BTBismack Tokoli
Bismack Tokoli
AAAbdulaziz Alhamadani
Abdulaziz Alhamadani
Airports play a vital role in global travel, serving millions of passengers daily. With more people traveling by air, airport service quality improvement is more important than ever to ensure continued passenger satisfaction and effective airport operation. Traditional methods for assessing Airport Service Quality (ASQ), such as structured surveys and sentiment analysis, often fail to capture the full spectrum of passenger experiences. Existing studies rely on sentiment scoring and topic modelling, which provide broad insights but lack the depth to identify specific service issues. This research proposes a novel approach leveraging Large Language Models (LLMs) to automate the analysis of airport customer reviews. Our methodology involves a four-phase process: (1) identification of extreme positive and extreme negative reviews using clustering-based methods on the sentiment scores and customer ratings, (2) keyword extraction using LLMs, (3) keywords categorization and (4) topics generation. The proposed method is applied to the Skytrax airport reviews dataset. With the combination of LLM-based keyword extraction and topic modelling, our system gives a richer understanding of passengers' experience in the form of overt concerns such as security delays, unclean facilities, or unfriendly staff interaction. This study aims to help airports make better decisions and improve customer experiences. Our findings demonstrate that LLMs can significantly enhance the granularity and relevance of ASQ insights compared to traditional methods. The proposed framework presents a scalable, automated solution for airport operators to systematically improve passenger experience and operational efficiency based on unstructured feedback.

25 January 2025
Verified
Book Chapter
Closed access
PDF file
AKAadyant Khatri
Aadyant Khatri
NENicolas Egierski
Nicolas Egierski
APAshutosh Pochamreddy
Ashutosh Pochamreddy
AAAbdulaziz Alhamadani
Abdulaziz Alhamadani
SSShailik Sarkar
Shailik Sarkar
CLChang-Tien Lu
Chang-Tien Lu