Digital Commons Network^™

Trustworthy Federated Learning Framework For Secure, Efficient, And Quality-Aware Distributed Ai, Asadullah Tariq

Dissertations

Federated Learning (FL) emerged as a significant advancement in the field of Artificial Intelligence (AI), enabling collaborative model training across distributed devices while maintaining data privacy. As the importance of FL and its application in various areas increased, addressing trustworthiness issues in its various aspects became crucial. In the FL process, clients contribute updates computed on their local datasets, which the server aggregates to iteratively refine the global model. However, not all client data may be relevant to the learning objective, and incorporating updates from irrelevant data can harm the model's performance. The selection of training samples significantly impacts model …

Startup Success Forecasting Through Machine Learning: A Comprehensive Analysis Of It Startups, Khaled Abdulla Alhassani

Theses

Lately, startups attracted significant attention from investors throughout the previous years. This raised several questions concerning startups and what they possibly define as them. It could refer to collective individuals who focus on innovative ideas with a reproducible and scalable business model; others refer to it as a newly established business. Nevertheless, all these definitions lead to a predictive question. Will these startups face success?

This study explores startup success prediction methods, focusing on forecasting information technology startup (SIT) insights using Machine Learning (ML) models such as Random Forest (RF), Decision Tree (DT), Support Vector Machine (SVM), K-Nearest Neighbor (k-NN), …

Explainable Sentiment Analysis With Deepseek-R1: Performance, Efficiency, And Few-Shot Learning, Donghao Huang, Zhaoxia Wang

Research Collection School Of Computing and Information Systems

Large language models (LLMs) have transformed sentiment analysis, yet balancing accuracy, efficiency, and explainability remains a critical challenge. This study presents the first comprehensive evaluation of DeepSeek-R1—an open-source reasoning model—against OpenAI’s GPT-4o and GPT-4o-mini. We test the full 671B model and its distilled variants, systematically documenting few-shot learning curves. Our experiments show DeepSeek-R1 achieves a 91.39% F1 score on 5-class sentiment and 99.31% accuracy on binary tasks with just 5 shots, an eightfold improvement in few-shot efficiency over GPT-4o. Architecture-specific distillation effects emerge, where a 32B Qwen2.5-based model outperforms the 70B Llama-based variant by 6.69 percentage points. While its reasoning …

Usefulness And Diminishing Returns: Evaluating Social Information In Recommender Systems, Qing Meng, Huiyu Min, Ming Shan Hee, Roy Ka-Wei Lee, Bing Tian Dai, Shuai Xu

Research Collection School Of Computing and Information Systems

Social recommendation, which leverages users’ social information to predict users’ preferences, is a popular branch of recommender systems. Many existing studies have attempted to advance the performance of collaborative filtering methods by leveraging the user-user matrix to enhance user embedding learning with user’s social connections. While the existing social recommender systems have demonstrated good performance in various recommendation tasks, the extent of social information usefulness in recommender systems remains unclear. This paper addresses the research gap by designing experiments to answer three research questions: (i) How useful is social information in varying user-item data sparsity? (ii) How much social information …

Metacan: Improving Generalizability Of Few‑Shot Anomaly Detection With Meta‑Learning, Zhisheng Lv, Jianfeng Zhang, Songlei Jian, Chenlin Huang, Hongguang Zhang, Guansong Pang, Zhong Liu

Research Collection School Of Computing and Information Systems

Few-shot Anomaly Detection (AD) for images aims to detect anomalies with few-shot normal samples from the target dataset. It is a crucial task when only few samples can be obtained, and it is challenging since it needs to be generalized to different domains. Existing methods try to enhance the generalizability of AD by incorporating large vision-language models (LVLMs).However, how to transform category semantic information in LVLMs into anomaly information to improve the generalizability of AD remains a challenge facing existing methods.To address the challenge, we propose a few-shot AD method called MetaCAN, a novel category-to-anomaly network trained with AD meta-learning …

International Workshop On Multimodal Generative Search And Recommendation (Mmgensr@Cikm 2025), Yi Bin, Haoxuan Li, Haokai Ma, Yang Zhang, Wenjie Wang, Yunshan Ma, Yang Yang, Tat‑Seng Chua

Research Collection School Of Computing and Information Systems

Recent breakthroughs in generative Artificial Intelligence (AI) have ignited a revolutionary wave across information retrieval and recommender systems. This workshop serves as a premier interdisciplinary platform to explore how generative models, particularly Large Language Models (LLMs) and Large Multimodal Models (LMMs), are transforming multimodal search and recommendation paradigms [3, 6, 9, 10, 12-14]. We aim to convene researchers and practitioners to discuss innovative architectures, methodologies, and evaluation strategies spanning generative document retrieval [5, 8] generative image retrieval [ 7, 16], grounded answer generation [17], generative recommendation [2, 4, 11], and related tasks involving multiple modalities [1,15]. The workshop will facilitate …

Ai Companions And The Lessons Of Family Law, Clare Huntington

Faculty Scholarship

Virtual friends and lovers powered by artificial intelligence are rapidly moving to the center of our emotional and social lives. Millions of people turn to AI companions every day for conversation, romance, sexual intimacy, therapy, and education. AI companionship holds promise, potentially reducing loneliness, supporting people without access to mental health treatment, helping students learn, and offering a judgment-free space for sensitive conversations. But AI companionship also raises significant concerns. The technology's addictiveness may exacerbate loneliness and can undermine human relationships. Therapy bots may prove more harmful than helpful. AI companions can be emotionally abusive. And their access to the …

Persepsi Mahasiswa Ilmu Perpustakaan Terhadap Penggunaan Perangkat Ai Llm Dalam Pencarian Informasi, Danisya Laila Zahra, Muhamad Prabu Wibowo

Jurnal Ilmu Informasi, Perpustakaan, dan Kearsipan

The increasing use of generative artificial intelligence (AI) among university students is driving changes in the way they seek and manage information, including in academic contexts. ChatGPT and DeepSeek AI are two AI platforms based on Large Language Models (LLMs) that are increasingly utilized as tools to support information seeking processes. This study aims to analyze the preferences of students from the Library and Information Science Program, Faculty of Humanities, Universitas Indonesia (FIB UI), in using these two platforms. The research employs a case study method with a qualitative approach, involving in-depth interviews with ten students. This study explores their …

Mosquito Classification And Explainability From Image Data Via Deep Learning Techniques, Farhat Binte Azam

USF Tampa Graduate Theses and Dissertations

According to the World Health Organization (WHO), mosquitoes are the deadliest animals on Earth, responsible for more human deaths annually than any other species. Mosquito-borne illnesses continue to pose severe risks to global health. In 2015 alone, there were an estimated 214 million malaria cases worldwide. Similarly, a 2016 report from the Centers for Disease Control and Prevention (CDC) revealed that Puerto Rico’s Department of Health received over 62,500 suspected cases of Zika, with 29,345 confirmed positive cases. In 2019, Southeast Asia experienced its worst dengue outbreak in recorded history. Of the approximately 4,500 mosquito species distributed across 34 genera, …

Applying Machine Learning Methods To Laser Acceleration Of Protons: Synthetic Data For Exploring The High Repetition Rate Regime, John J. Felice, Ronak Desai, Nathaniel Tamminga, Joseph R. Smith, Alona Kryshchenko, Christopher M. Orban, Michael L. Dexter, Anil K. Patnaik

Faculty Publications

Advances in ultra‐intense laser technology have increased repetition rates and average power for chirped‐pulse laser systems, which offer a promising solution for many applications including energetic proton sources. An important challenge is the need to optimize and control the proton source by varying some of the many degrees of freedom inherent to the laser‐plasma interactions. Machine learning can play an important role in this task, as our work examines. Building on our earlier work in Desai et al. 2024, we generate a large ∼1.5 million data point synthetic data set for proton acceleration using a physics‐informed analytic model that we …

Attention Mapping For Hallucination Reduction In Arabic Financial Using Retrieval-Augmented Generation Systems, Hasan Abdulameer Hasan, Khaldoun H. Al-Hussayni, Ali Z. K. Matloob

Journal of Intelligent Informatics, Networking, and Cybersecurity

This research introduces a customized attention visualization framework for mitigating hallucinations in Arabic Retrieval-Augmented Generation (RAG) systems tailored for financial document analysis. The proposed architecture extends MarBERT with a dual-stage attention supervision mechanism and a hallucination-aware loss formulation, trained on a newly constructed dataset of 7,000 annotated Arabic financial query-context pairs. A grounding alignment score is computed over attended tokens, and generated responses are rejected when falling below a dynamically adjusted precision-aware threshold,serving as the core decision-making approach for hallucination detection. The system achieves 95.04% classification accuracy, 95.84% precision, 94.12% recall, and an F1 score of 94.97%, outperforming AraELECTRA and …

Higher Education Cybersecurity: A Vulnerability Assessment Of The U.S. South’S Institutional Websites, Zachary W. Taylor, Vivi Vo

Journal of Cybersecurity Education, Research and Practice

As technology continues to advance, it is critical to understand how higher education institutions protect digital information of their stakeholders including students, faculty, and staff through cybersecurity measures. Although conceptual research has articulated various aspects of cybersecurity, no empirical research has explored the cybersecurity of higher education (.edu) websites through a vulnerability scan of these websites via an open PortScan and analysis. To fill a critical gap in the literature, this study conducted a vulnerability scan and open PortScan and analysis of all higher education websites in three of the lowest-income states in the United States: Louisiana (n=112), Mississippi (n=52), …

Efficient Smooth Tensor Train And Tensor Ring Completion For Image Classification Enhancement, Salman Ahmadi-Asl, Roman V. Garaev, Rustam A. Lukmanov, Naeim Rezaeian, Asad Masood Khattak, Manuel Mazzara

All Works

This paper deals with studying the data completion problem for enhancing the image classification task under the pixel removal scenario. In some applications, it happens that a part of the pixels of a given image is lost due to several issues, such as corruption by outliers or artifacts and/or incompleteness due to imprecise data acquisition. This issue results in a completely wrong classification outcome using Deep Neural Networks (DNNs). In this paper we investigate the benefit of data completion in enhancing the classification accuracy of the DNN models to build more robust and stable DNN models. To this end, we …

Cross-Model Watermarking Via Discriminative Samples For Secure Authentication, Juan Zhao, Yudao Sun, Zhihai Yang, Cai Xu, Hongji Chen, Fan Zhang, Jianxin Li

Research outputs 2022 to 2026

Deep neural networks on cloud platforms face growing security threats, with AI services increasingly relying on heterogeneous models for the same task to meet diverse user needs. Existing methods fail to distinguish benign modifications from malicious attacks in cross-model scenarios. To address this challenge, we propose a non-intrusive cross-model watermarking method that generates discriminative samples as universal keys, enabling authentication without altering model parameters or architectures. Specifically, we introduce a margin enhancement loss to amplify confidence gaps between benign and malicious behaviors, ensuring high transferability across models. Both theoretical analysis and experimental results demonstrate the high efficacy of our proposed …

Bimw: Blockchain-Enabled Innocuous Model Watermarking For Secure Ownership Verification, Xinyun Liu, Ronghua Xu

Michigan Tech Publications

The integration of artificial intelligence (AI) and edge computing gives rise to edge intelligence (EI), which offers effective solutions to the limitations of traditional cloud-based AI; however, deploying models across distributed edge platforms raises concerns regarding authenticity, thereby necessitating robust mechanisms for ownership verification. Currently, backdoor-based model watermarking techniques represent a state-of-the-art approach for ownership verification; however, their reliance on model poisoning introduces potential security risks and unintended behaviors. To solve this challenge, we propose BIMW, a blockchain-enabled innocuous model watermarking framework that ensures secure and trustworthy AI model deployment and sharing in distributed edge computing environments. Unlike widely applied …

User Privacy In The Digital Playground: An In-Depth Investigation Of Facebook Instant Games, Sideeq Bello

LSU Master's Theses

Amid growing concerns over data privacy in web and mobile applications, this study aims to assess the privacy mechanisms in instant games on Facebook, a platform with approximately 3.03 billion monthly active users and a substantial repository of personal data. Instant Games have become increasingly popular due to their ease of access and social integration features. Investigating these games can provide insights into privacy mechanisms and practices, thereby informing the development of more fair, compliant, and user privacy-centric gaming experiences. Thus, this study proposes an integrated analytical framework that leverages a combination of descriptive, memory, and network analysis techniques to …

Neutrosophic Decision Making Methodology For Sustainable Diabetic Diet: Balancing Health And Environment, Rayan Hussein

Neutrosophic Systems with Applications

Sustainable diabetic diet for balancing health and environment is a decision-making problem and contain uncertainty affect negatively in different decisions. So, this paper proposes a decision-making methodology with the neutrosophic set to solve uncertainty in this decision-making problem. We use the MOOSRA method to rank the alternatives. The criteria weights are computed using the average method. Case study with nine criteria and 16 alternatives are proposed for sustainable diabetic diet. The sensitivity analysis is conducted in this case study to show the stability of the ranks. Nine cases are proposed in this study. The results show the ranks of alternatives …

Neutrosophic Algebraic Structures For Precise Uncertainty Quantification In Outcome-Based Education Systems: A Rigorous Case Study Analysis, Mona Gharib, Imran Siddique, Miin Shen Yang

Neutrosophic Systems with Applications

Outcome-Based Education (OBE) emphasizes measurable learning results, yet the evaluation of professional talent training in physical education often involves uncertain, incomplete, or even contradictory indicators. Traditional assessment models are limited in capturing these indeterminacies. To address this challenge, we propose a novel neutrosophic algebraic framework that integrates neutrosophic probability, measure, and algebraic structures with OBE evaluation. This study introduces a neutrosophic evaluation framework that captures both determinate and indeterminate aspects of brand competitiveness. A case study on physical education professional training demonstrates how the proposed model captures hidden uncertainty and provides a more balanced assessment than classical methods. The results …

Neutrosophic Set Model For Effective Earthquake Disaster Risk Management: Results And Discussion, Emadaldeen Hassan Alomar, Abdullah Ali Salamai

Neutrosophic Systems with Applications

Secondary effects including landslides, tsunamis, and fires can cause significant damage and fatalities following an earthquake. Effective disaster risk management has been predicted to be built on regional fire-following earthquake (FFE) risk. Specifically, a target region’s building and geographical factors might impact the fire danger and spread. The percentage of fire-resistant building types as building characteristics and the distribution of building densities as regional characteristics were the primary factors used in this study to determine FFE risk. This study develops a decision-making methodology for risk management in the FFE. We use the single valued neutrosophic set (SVNS) to overcome uncertainty. …

Evaluation Of Soil Quality Index For Sugar Beet Cultivation Under Neutrosophic Set Framework, Asiye Yilmaz Adkinson

Neutrosophic Systems with Applications

Evaluation of soil quality index for sugar beet cultivation is a decision-making problem. So, we use the decision-making methods to solve this problem. We use the ARAS method is a decision-making method to rank the alternatives. The average method is used to compute the criteria weights. Evaluation this problem contains uncertainty information. So, the neutrosophic set is used to solve this uncertainty information. This study uses ten criteria and 15 alternatives. The sensitivity analysis is conducted to show the stability of the ranks. We change the criteria weights by ten cases. Then we apply the steps of the proposed approach …

Sustainable Assessing Cross-Border Renewable Energy Alliances Using Neutrosophic Numbers With Long-Term Energy Transition Planning, Kamal Alieyan, Amr A. Abd El-Mageed

Neutrosophic Systems with Applications

This paper proposes a decision-making methodology for Sustainable assessing cross-border renewable energy alliances. We used two decision-making methods such as Entropy and MABAC methods. Entropy method is used to compute the criteria weights. The MABAC method is used to rank the alternatives. Two methods are used under the neutrosophic number to solve uncertainty in the decision making. Three stages of the proposed approach are conducted. In the first stage, we compute the criteria weights. In the second stage, we rank the alternatives. In the third stage, we conducted the sensitivity analysis to show the stability of the ranks. The results …

Enhancing Cyberattack Resiliency Through The Radiotherapy Backup And Recovery Dashboard Tool, Justin Pijanowski, Eric Nguyen, Yasin Abdulkadir, Justin Hink, Yevgeniy Vinogradskiy, James Lamb

Department of Radiation Oncology Faculty Papers

PURPOSE: Radiation Oncology departments impacted by recent cyberattacks were unable to access data backups or their Record and Verify (R&V) system and therefore faced challenges to resume patient treatments in a timely manner. We present a novel software tool that backs-up critical radiotherapy treatment information and displays essential information for on-treatment patients in an intuitive and accessible dashboard allowing clinics to continue radiotherapy treatments. The purpose of this report is to describe implementation details, challenges, and share open-source code to facilitate radiation oncology clinics' efforts to develop tools to improve cyberattack resiliency.

METHODS: The Radiotherapy Backup and Recovery Dashboard Tool …

Myfoodrx: A Personalized Food-As-Medicine Mhealth Application For Food-Insecure Adults With Chronic Conditions, Jay Hiteshkumar Jariwala

USF Tampa Graduate Theses and Dissertations

Food insecurity (FI) remains a persistent public health challenge in the United States, disproportionately affecting underserved populations and contributing to higher rates of chronic conditions such as diabetes, hypertension, and obesity. Traditional Food-as-Medicine programs have emerged as promising interventions, but often follow a generalized, non-personalized approach that limits long-term effectiveness, especially among diverse, high-risk communities. This thesis presents MyFoodRx, a personalized mobile health (mHealth) application designed to address the intersection of food insecurity and chronic disease through tailored nutritional guidance, real-time pantry integration, and adaptive educational content.

Developed through a User-Centered Design (UCD) process, MyFoodRx leverages a modular client–server architecture, …

Towards Mitigation Of Hallucination For Llm-Empowered Agents: Progressive Generalization Bound Exploration And Watchdog Monitor, Siyuan Liu, Wenjing Liu, Zhiwei Xu, Xin Wang, Bo Chen, Tao Li

Michigan Tech Publications

Empowered by large language models (LLMs), intelligent agents have become a popular paradigm for interacting with open environments to facilitate AI deployment. However, hallucinations generated by LLMs - where outputs are inconsistent with facts - pose a significant challenge, undermining the credibility of intelligent agents. Only if hallucinations can be mitigated, the intelligent agents can be used in real-world without any catastrophic risk. Therefore, effective detection and mitigation of hallucinations are crucial to ensure the dependability of agents. Unfortunately, the related approaches either depend on white-box access to LLMs or fail to accurately identify hallucinations. To address the challenge posed …

The Orange Glow In The Sunshine State: Three Visions Of Plato In Florida (1970-1990), Ryan Mcgahan

USF Tampa Graduate Theses and Dissertations

The PLATO network, a collection of mainframe computers, terminals, and educational software, has received an increasing amount of scholarly attention in the last decade as a social precursor to the modern internet. Existing histories have neglected to investigate the ways in which PLATO and its related business enterprise worked to accelerate the shift in university governance away from classical liberal ideas centering the public good and towards a more profit-centered neoliberal rationality. By tracing the rise of PLATO in Florida universities, this paper argues that PLATO aided and was aided by the shifting priorities of American universities in the 1970s …

Inferred Global Dense Residue Transition Graphs From Primary Structure Sequences Enable Protein Interaction Prediction Via Directed Graph Convolutional Neural Networks, Islam A. Ebeid, Haoteng Tang, Pengfei Gu

Computer Science Faculty Publications

Introduction: Accurate prediction of protein-protein interactions (PPIs) is crucial for understanding cellular functions and advancing the development of drugs. While existing in-silico methods leverage direct sequence embeddings from Protein Language Models (PLMs) or apply Graph Neural Networks (GNNs) to 3D protein structures, the main focus of this study is to investigate less computationally intensive alternatives. This work introduces a novel framework for the downstream task of PPI prediction via link prediction.

Methods: We introduce a two-stage graph representation learning framework, ProtGram-DirectGCN. First, we developed ProtGram, a novel approach that models a protein's primary structure as a hierarchy of …

An Empirical Investigation Into The Effect Of Brain Drain “Japa Syndrome” On Agile Practitioners Developing Healthcare Information Systems Software In Nigeria, Yazidu Buba Salihu, Julian M. Bass, Gloria E. Iyawa

Communications of the IIMA

Within the software engineering context, agile approaches encourage customer collaboration, iterative development, and flexibility. However, the mass exodus of highly skilled professionals, known as the “brain drain” or "Japa Syndrome," has emerged as a significant challenge, especially in Nigeria. This phenomenon has particularly impacted agile software development practitioners by undermining project continuity, knowledge transfer, and team dynamics. This study empirically examines the effect of brain drain, “Japa Syndrome,” on agile practitioners developing healthcare information systems software in Nigeria. It employed a qualitative, multi-method approach to gather empirical data from 13 agile practitioners in Nigeria’s healthcare information systems sector. The study …

Multisource Information Fusion Method For Human Gait Perception, Guiliang Chen, Guowei Liu, Yongchao Li, Chao Cai, Zihao Li, Dong Yang

Journal of System Simulation

Abstract: In response to the insufficient gait perception capability during lower limb exoskeleton assistance, a human lower limb gait phase optimization classification model was proposed. A wireless transmission gait information collection system was designed for collecting the required gait phase feature information. Human joint angles were accurately calculated by fusing acceleration and angular velocity information using extended Kalman filtering. Additionally, kernel principal component analysis was applied to reduce dimensionality in conjunction with plantar pressure data. The LSSVM algorithm was employed to classify gait data, and the PSO algorithm was utilized to find the optimal classification parameters. Experimental results demonstrate that …

Optimization Dispatch Method For High-Proportion Renewable Energy Power Systems Based On Sc-Ppo, Zhongkai Xu, Chenyang Chu, Kai Xie, Ruizhuo Zhao, Wenjun Ke

Journal of System Simulation

Abstract: The high proportion of renewable energy integration brings significant challenges of randomness, multi-objective coupling, and security constraints to power systems. Traditional model-driven methods have limitations in modeling accuracy and adaptability. To address these issues, this paper proposed a safety-constrained PPO algorithm (SC-PPO). The method included three improvements. A temporal convolutional network was utilized to construct a dynamic state encoder that integrated historical operation, real-time monitoring, and prediction data to form a causal state representation. A hierarchical reward structure was designed, and an adaptive weighting mechanism based on constraint satisfaction degree was introduced to coordinate multi-objective optimization. Physical constraint projection …

Impact Of Information Security Awareness Training On Knowledge, Attitude, And Behavior: A K-12 Case Study, Michael S. Robbins, Christopher Robbins

Journal of Cybersecurity Education, Research and Practice

Abstract— Information security breaches remain a serious threat across all sectors, often exploiting human factors rather than technical flaws. This study examines how a structured Information Security Awareness (ISA) training program influences employees’ knowledge of security policies, attitudes towards those policies, and self-reported security behaviors within a K-12 educational environment. A quantitative pre-test/post-test design was employed with 201 staff members (administrators, teachers, and support personnel) in a public school district. Participants completed the Human Aspects of Information Security Questionnaire (HAIS-Q) before and after undergoing an interactive cybersecurity training program. Statistical analysis revealed a significant improvement in information security knowledge, attitudes, …