Physical Sciences and Mathematics Commons^™

Analysis Of Green Data Center Efforts And Energy Usage, Dillon J. Goicoechea

Honors Projects

This paper is an undergraduate level literature review and analysis of research surrounding the Green Data Center phenomenon. Review of work covering energy usage, data usage, usage predictions, and strategies for decreasing energy requirements is the main analysis of this work. The analysis shows that while data centers are becoming greener, the increase in usage of their capacities is negating those efficiency increases. The increase in the energy efficiency of data centers is crucial, however, there must be made efforts to lower computational and data usage to help achieve lower energy usage of data centers.

Building A Data Pipeline And Machine Learning Model For Insurance Data, Connor Weyers

Honors Theses

Insurance telematics is an emerging and exciting field. It combines the advancements in GPS tracking, computational analytics, data processing, and machine learning into a useful tool to help insurance companies make the best product for their consumers. This is why National Indemnity looked to implement a telematics portion to their business processes of underwriting insurance policies and sponsored a School of Computing Senior Design project. In this report, we will first review existing solutions that been used to solve problems and subproblems similar to that we are given in this project. We then propose designs for the data pipeline and …

Online Temporal Data Mining And Learning: Pursuing Enhanced Efficiency And Robust Algorithms, Sheng Zhong

Computer Science ETDs

Time series data mining and learning serve as a cornerstone across various domains, including finance, healthcare, and science. Recent advancements in network and sensor technologies have ignited an increasing interest in real-time temporal data mining and learning techniques. Various tasks benefit from these techniques, such as environmental monitoring, event detection, anomaly identification, and forecasting. However, these techniques still face significant challenges in the online environment settings, encompassing aspects like efficiency, accuracy, robustness, and scarcity of labeled data. This dissertation presents four innovative solutions: FilCorr, DCT-MASS, FewSig, and BitLINK to overcome these challenges. We evaluate each method and showcase their practical …

Machine Learning: Face Recognition, Mohammed E. Amin

Publications and Research

This project explores the cutting-edge intersection of machine learning (ML) and face recognition (FR) technology, utilizing the OpenCV library to pioneer innovative applications in real-time security and user interface enhancement. By processing live video feeds, our system encodes visual inputs and employs advanced face recognition algorithms to accurately identify individuals from a database of photos. This integration of machine learning with OpenCV not only showcases the potential for bolstering security systems but also enriches user experiences across various technological platforms. Through a meticulous examination of unique facial features and the application of sophisticated ML algorithms and neural networks, our project …

Improving Tattle-Tale K-Deniability, Nicholas G.E. Morales

Student Research Symposium

Ensuring privacy for databases is an ongoing struggle. While the majority of work has focused on using access control lists to protect sensitive data these methods are vulnerable to inference attacks. A set of algorithms, referred to as Tattle-Tale, was developed that could protect sensitive data from being inferred however its runtime performance wasn’t suitable for production code. This set of algorithms contained two main subsets, Full Deniability and K-Deniability. My research focused on improving the runtime or utility of the K-Deniability algorithms. I investigated the runtime of the K-Deniability algorithms to identify what was slowing the process down. Aside …

A Novel Caching Algorithm For Efficient Fine-Grained Access Control In Database Management Systems, Anadi Shakya

Student Research Symposium

Fine-grained access Control (FGAC) in DBMS is vital for restricting user access to authorized data and enhancing security. FGAC policies govern how users are granted access to specific resources based on detailed criteria, ensuring security and privacy measures. Traditional methods struggle with scaling policies to thousands, causing delays in query responses. This paper introduces a novel caching algorithm designed to address this challenge by accelerating query processing and ensuring compliance with FGAC policies. In our approach, we create a circular hashmap and employ different replacement techniques to efficiently manage the cache, prioritizing entries that are visited more frequently. To evaluate …

Identification Of Conceptual Neighborhoods And Topological Relations In Z2, Brendan P. Hall

Electronic Theses and Dissertations

Topological relations are an essential element of spatial queries and reasoning about spatial information. The predominant model for topological relations in geographic information systems—the 9-intersection—identifies sixteen different relations between groups of pixels (called raster regions) given a set of conditions restricting the composition of the regions interior and boundary. Several of these relations are dependent on the raster region sizes to be realized. An example, ‘Completely Inside' would require raster regions to be sufficiently different in size for one raster to entirely encompass the other. By developing an iterative computational model, this work generates conceptual neighborhood graphs that outlined …

Code For Care: Hypertension Prediction In Women Aged 18-39 Years, Kruti Sheth

Electronic Theses, Projects, and Dissertations

The longstanding prevalence of hypertension, often undiagnosed, poses significant risks of severe chronic and cardiovascular complications if left untreated. This study investigated the causes and underlying risks of hypertension in females aged between 18-39 years. The research questions were: (Q1.) What factors affect the occurrence of hypertension in females aged 18-39 years? (Q2.) What machine learning algorithms are suited for effectively predicting hypertension? (Q3.) How can SHAP values be leveraged to analyze the factors from model outputs? The findings are: (Q1.) Performing Feature selection using binary classification Logistic regression algorithm reveals an array of 30 most influential factors at an …

Database And Machine Learning Model For Classifying Autism Spectrum Disorder From Smartphone Based Electroretinography, Rory Harris

Honors Scholar Theses

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder that negatively affects a patient’s cognitive and communication aptitude and, therefore, can severely impact that patient’s quality of life. Because of this, early diagnosis is paramount. In recent studies, electroretinography (ERG), which is a measure of the retina’s electrical response to a brief flash of light into the eye, has shown promise in detecting ASD. Access to these scans can provide early diagnosis, improving well-being. Current ERG devices are very expensive due to their on board processing capabilities. This paper aims to create an ERG device using a smartphone as the main …

The Quantitative Analysis And Visualization Of Nfl Passing Routes, Sandeep Chitturi

Computer Science and Computer Engineering Undergraduate Honors Theses

The strategic planning of offensive passing plays in the NFL incorporates numerous variables, including defensive coverages, player positioning, historical data, etc. This project develops an application using an analytical framework and an interactive model to simulate and visualize an NFL offense's passing strategy under varying conditions. Using R-programming and data management, the model dynamically represents potential passing routes in response to different defensive schemes. The system architecture integrates data from historical NFL league years to generate quantified route scores through designed mathematical equations. This allows for the prediction of potential passing routes for offensive skill players in response to the …

Improving The Performance Of Wi-Fi Indoor Localization In Both Dense And Unknown Environments, Quang Truong Hai

Dissertations and Theses Collection (Open Access)

Indoor localization is important for various pervasive applications, garnering considerable research attention over recent decades. Despite numerous proposed solutions, the practical application of these methods in real-world environments with high applicability remains challenging. One compelling use case for building owners is the ability to track individuals as they navigate through the building, whether for security, customer analytics, space utilization planning, or other management purposes. However, this task becomes exceedingly difficult in environments with hundreds or thousands of people in motion. Conversely, the need to track oneself’s location is also meaningful from the perspective of individuals traversing in crowded spaces. These …

Cybervictimization And Depression: A Cultural Standpoint, Paige Heagy

University Honors College

The author’s aim was to investigate the relationship between cybervictimization and depression, as well as using peer attachment and culture as moderators by giving questionnaires to 1347 participants (age range = 11-15 years) from India and the United States. Through four questionnaires, adolescents reported their levels of endorsement in either individualism or collectivism culture, levels of cybervictimization, levels of depression/depressive symptoms, and their levels of peer attachment. Adolescents reported that there is a significantly positive correlation between cybervictimization and depression. Differences were found according to culture and peer attachment, as well. Cybervictimization has begun to erupt worldwide as internet usage …

Multigprompt For Multi-Task Pre-Training And Prompting On Graphs, Xingtong Yu, Chang Zhou, Yuan Fang, Xinming Zhan

Research Collection School Of Computing and Information Systems

Graph Neural Networks (GNNs) have emerged as a mainstream technique for graph representation learning. However, their efficacy within an end-to-end supervised framework is significantly tied to the availability of task-specific labels. To mitigate labeling costs and enhance robustness in few-shot settings, pre-training on self-supervised tasks has emerged as a promising method, while prompting has been proposed to further narrow the objective gap between pretext and downstream tasks. Although there has been some initial exploration of prompt-based learning on graphs, they primarily leverage a single pretext task, resulting in a limited subset of general knowledge that could be learned from the …

Plug-And-Play Policy Planner For Large Language Model Powered Dialogue Agents, Yang Deng, Wenxuan Zhang, Wai Lam, See-Kiong Ng, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Proactive dialogues serve as a practical yet challenging dialogue problem in the era of large language models (LLMs), where the dialogue policy planning is the key to improving the proactivity of LLMs. Most existing studies enable the dialogue policy planning of LLMs using various prompting schemes or iteratively enhance this capability in handling the given case with verbal AI feedback. However, these approaches are either bounded by the policy planning capability of the frozen LLMs or hard to be transferred to new cases. In this work, we introduce a new dialogue policy planning paradigm to strategize LLMs for proactive dialogue …

Unraveling The ‘Anomaly’ In Time Series Anomaly Detection: A Self-Supervised Tri-Domain Solution, Yuting Sun, Guansong Pang, Guanhua Ye, Tong Chen, Xia Hu, Hongzhi Yin

Research Collection School Of Computing and Information Systems

The ongoing challenges in time series anomaly detection (TSAD), including the scarcity of anomaly labels and the variability in anomaly lengths and shapes, have led to the need for a more robust and efficient solution. As limited anomaly labels hinder traditional supervised models in anomaly detection, various state-of-the-art (SOTA) deep learning (DL) techniques (e.g., self-supervised learning) are introduced to tackle this issue. However, they encounter difficulties handling variations in anomaly lengths and shapes, limiting their adaptability to diverse anomalies. Additionally, many benchmark datasets suffer from the problem of having explicit anomalies that even random functions can detect. This problem is …

Diffusion-Based Negative Sampling On Graphs For Link Prediction, Yuan Fang, Yuan Fang

Research Collection School Of Computing and Information Systems

Link prediction is a fundamental task for graph analysis with important applications on the Web, such as social network analysis and recommendation systems, etc. Modern graph link prediction methods often employ a contrastive approach to learn robust node representations, where negative sampling is pivotal. Typical negative sampling methods aim to retrieve hard examples based on either predefined heuristics or automatic adversarial approaches, which might be inflexible or difficult to control. Furthermore, in the context of link prediction, most previous methods sample negative nodes from existing substructures of the graph, missing out on potentially more optimal samples in the latent space. …

On The Feasibility Of Simple Transformer For Dynamic Graph Modeling, Yuxia Wu, Yuan Fang, Lizi Liao

Research Collection School Of Computing and Information Systems

Dynamic graph modeling is crucial for understanding complex structures in web graphs, spanning applications in social networks, recommender systems, and more. Most existing methods primarily emphasize structural dependencies and their temporal changes. However, these approaches often overlook detailed temporal aspects or struggle with long-term dependencies. Furthermore, many solutions overly complicate the process by emphasizing intricate module designs to capture dynamic evolutions. In this work, we harness the strength of the Transformer’s self-attention mechanism, known for adeptly handling long-range dependencies in sequence modeling. Our approach offers a simple Transformer model, called SimpleDyG, tailored for dynamic graph modeling without complex modifications. We …

Dlvs4audio2sheet: Deep Learning-Based Vocal Separation For Audio Into Music Sheet Conversion, Nicole Teo, Zhaoxia Wang, Ezekiel Ghe, Yee Sen Tan, Kevan Oktavio, Alexander Vincent Lewi, Allyne Zhang, Seng-Beng Ho

Research Collection School Of Computing and Information Systems

While manual transcription tools exist, music enthusiasts, including amateur singers, still encounter challenges when transcribing performances into sheet music. This paper addresses the complex task of translating music audio into music sheets, particularly challenging in the intricate field of choral arrangements where multiple voices intertwine. We propose DLVS4Audio2Sheet, a novel method leveraging advanced deep learning models, Open-Unmix and Band-Split Recurrent Neural Networks (BSRNN), for vocal separation. DLVS4Audio2Sheet segments choral audio into individual vocal sections and selects the optimal model for further processing, aiming towards audio into music sheet conversion. We evaluate DLVS4Audio2Sheet’s performance using these deep learning algorithms and assess …

Learning Multi-Faceted Prototypical User Interests, Nhu Thuat Tran, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

We seek to uncover the latent interest units from behavioral data to better learn user preferences under the VAE framework. Existing practices tend to ignore the multiple facets of item characteristics, which may not capture it at appropriate granularity. Moreover, current studies equate the granularity of item space to that of user interests, which we postulate is not ideal as user interests would likely map to a small subset of item space. In addition, the compositionality of user interests has received inadequate attention, preventing the modeling of interactions between explanatory factors driving a user's decision. To resolve this, we propose …

Anomalyclip: Object-Agnostic Prompt Learning For Zero-Shot Anomaly Detection, Qihang Zhou, Guansong Pang, Yu Tian, Shibo He, Jiming Chen

Research Collection School Of Computing and Information Systems

Zero-shot anomaly detection (ZSAD) requires detection models trained using auxiliary data to detect anomalies without any training sample in a target dataset. It is a crucial task when training data is not accessible due to various concerns, e.g., data privacy, yet it is challenging since the models need to generalize to anomalies across different domains where the appearance of foreground objects, abnormal regions, and background features, such as defects/tumors on different products/ organs, can vary significantly. Recently large pre-trained vision-language models (VLMs), such as CLIP, have demonstrated strong zero-shot recognition ability in various vision tasks, including anomaly detection. However, their …

Non-Vacuous Generalization Bounds For Adversarial Risk In Stochastic Neural Networks, Mustafa Waleed, Liznerski Philipp, Antoine Ledent, Wagner Dennis, Wang Puyu, Kloft Marius

Research Collection School Of Computing and Information Systems

Adversarial examples are manipulated samples used to deceive machine learning models, posing a serious threat in safety-critical applications. Existing safety certificates for machine learning models are limited to individual input examples, failing to capture generalization to unseen data. To address this limitation, we propose novel generalization bounds based on the PAC-Bayesian and randomized smoothing frameworks, providing certificates that predict the model’s performance and robustness on unseen test samples based solely on the training data. We present an effective procedure to train and compute the first non-vacuous generalization bounds for neural networks in adversarial settings. Experimental results on the widely recognized …

Compositional Policy Learning In Stochastic Control Systems With Formal Guarantees, Dorde Zikelic, Mathias Lechner, Abhinav Verma, Krishnendu Chatterjee, Thomas A. Henzinger

Research Collection School Of Computing and Information Systems

Reinforcement learning has shown promising results in learning neural network policies for complicated control tasks. However, the lack of formal guarantees about the behavior of such policies remains an impediment to their deployment. We propose a novel method for learning a composition of neural network policies in stochastic environments, along with a formal certificate which guarantees that a specification over the policy's behavior is satisfied with the desired probability. Unlike prior work on verifiable RL, our approach leverages the compositional nature of logical specifications provided in SPECTRL, to learn over graphs of probabilistic reach-avoid specifications. The formal guarantees are provided …

Evaluation Of Orca 2 Against Other Llms For Retrieval Augmented Generation, Donghao Huang, Zhaoxia Wang

Research Collection School Of Computing and Information Systems

This study presents a comprehensive evaluation of Microsoft Research’s Orca 2, a small yet potent language model, in the context of Retrieval Augmented Generation (RAG). The research involved comparing Orca 2 with other significant models such as Llama-2, GPT-3.5-Turbo, and GPT-4, particularly focusing on its application in RAG. Key metrics, included faithfulness, answer relevance, overall score, and inference speed, were assessed. Experiments conducted on high-specification PCs revealed Orca 2’s exceptional performance in generating high quality responses and its efficiency on consumer-grade GPUs, underscoring its potential for scalable RAG applications. This study highlights the pivotal role of smaller, efficient models like …

Large Language Model Powered Agents In The Web, Yang Deng, An Zhang, Yankai Lin, Xu Chen, Ji-Rong Wen, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Web applications serve as vital interfaces for users to access information, perform various tasks, and engage with content. Traditional web designs have predominantly focused on user interfaces and static experiences. With the advent of large language models (LLMs), there’s a paradigm shift as we integrate LLM-powered agents into these platforms. These agents bring forth crucial human capabilities like memory and planning to make them behave like humans in completing various tasks, effectively enhancing user engagement and offering tailored interactions in web applications. In this tutorial, we delve into the cutting-edge techniques of LLM-powered agents across various web applications, such as …

Multithreaded Applications On The Heterogeneous Research Computing Environment., Sungbo Jung

Electronic Theses and Dissertations

Bioinformatics is a domain that has experienced rapid research growth in recent years, as evidenced by the increasing number of articles in biomedical databases such as PubMed, which adds over a million publications every year. However, this also poses a challenge for researchers who need to find relevant citations for their work. Therefore, developing efficient indexing and searching methods for text data is crucial for Bioinformatics. One key technique for information retrieval is document inversion, which involves creating an inverted index to enable efficient searching through vast collections of text or documents. This Ph.D. research aims to design the research …

A Little Loud And A Little Alone: A Phenomenology Of Leadership Identity Construction Among Women In Higher Education Technology, Amy Barry

Department of Teaching, Learning, and Teacher Education: Theses and Other Student Research

This qualitative study is an exploration of how women in higher education information technology (IT) positions navigate constructing their leadership identities. This includes the messy, personal, internal identity work that occurs prior to claiming their leadership identities on the public stage, followed by an examination of what the experience of attempting to claim and negotiate a leadership identity is like in the social context of their organizations. This educational and sociological study employs an Interpretative Phenomenological Analysis approach with a series of three interviews per participant that allowed the researcher to deeply explore the personal identity experiences of participants. Findings …

Using Dynamic Schemas For Query Optimization Over Json Data., Tomas Felipe Llano-Rios

Electronic Theses and Dissertations

Query optimization in document stores has traditionally relied on rule-based approaches, but recent research advocates for a shift towards cost-based optimization. However, this transition is hindered by the fragmented nature of existing approaches, stemming from the early development stage of cost-based query optimization for document databases. A key challenge lies in the absence of a standardized query language and semantics, exacerbated by the diverse and schema-less nature of JSON document collections. To tackle these challenges, the literature has proposed dynamic schemas, primarily utilized at parsing time. However, these schemas lack a formal foundation that describes meaningful semantics for query optimization. …

Comparative Predictive Analysis Of Stock Performance In The Tech Sector, Asaad Sendi

University of New Orleans Theses and Dissertations

This study compares the performance of deep learning models, including Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Transformer, in predicting stock prices across five companies (AAPL, CSCO, META, MSFT, and TSLA) from July 2019 to July 2023. Key findings reveal that GRU models generally exhibit the lowest Mean Absolute Error (MAE), indicating higher precision, particularly notable for CSCO with a remarkably low MAE. While LSTM models often show slightly higher MAE values, they outperform Transformer models in capturing broader trends and variance in stock prices, as evidenced by higher R-squared (R2) values. Transformer models generally exhibit higher MAE …

An Empirical Study On The Efficacy Of Llm-Powered Chatbots In Basic Information Retrieval Tasks, Naja Faysal

Electronic Theses, Projects, and Dissertations

The rise of conversational user interfaces (CUIs) powered by large language models (LLMs) is transforming human-computer interaction. This study evaluates the efficacy of LLM-powered chatbots, trained on website data, compared to browsing websites for finding information about organizations across diverse sectors. A within-subjects experiment with 165 participants was conducted, involving similar information retrieval (IR) tasks using both websites (GUIs) and chatbots (CUIs). The research questions are: (Q1) Which interface helps users find information faster: LLM chatbots or websites? (Q2) Which interface helps users find more accurate information: LLM chatbots or websites?. The findings are: (Q1) Participants found information significantly faster …

Representation Learning For Generative Models With Applications To Healthcare, Astronautics, And Aviation, Van Minh Nguyen

Theses and Dissertations

This dissertation explores applications of representation learning and generative models to challenges in healthcare, astronautics, and aviation.

The first part investigates the use of Generative Adversarial Networks (GANs) to synthesize realistic electronic health record (EHR) data. An initial attempt at training a GAN on the MIMIC-IV dataset encountered stability and convergence issues, motivating a deeper study of 1-Lipschitz regularization techniques for Auxiliary Classifier GANs (AC-GANs). An extensive ablation study on the CIFAR-10 dataset found that Spectral Normalization is key for AC-GAN stability and performance, while Weight Clipping fails to converge without Spectral Normalization. Analysis of the training dynamics provided further …