Digital Commons Network^™

Mass-Ratio-Convex-Combination-Of-Average-Absolute-Deviation-And-Interquartile-Range Based Outlier Factor For Anomaly Scoring, Zehong Fan

Chulalongkorn University Theses and Dissertations (Chula ETD)

Anomaly detection is increasingly essential across various sectors, including fraud detection, cybersecurity, and industrial process monitoring, where it serves as an early warning system to identify unusual patterns in data. This research presents a novel, parameter-free anomaly scoring algorithm designed to assign scores to data points without the need for user-defined parameters. The algorithm calculates anomaly scores based on the statistical dispersion of the mass-ratio distribution, utilizing five key measures: variance, range, interquartile range (IQR), average absolute deviation (AAD), and the combination of AAD and IQR. The performance of the proposed Mass-Ratio-Convex-Combination-Of-Average-Absoulte-Deviation-And-Interquartile-Range Based Outlier Factor for anomaly scoring algorithm called …

Federated Graph Anomaly Detection Via Contrastive Self-Supervised Learning, Xiangjie Kong, Wenyi Zhang, Hui Wang, Mingliang Hou, Xin Chen, Xiaoran Yan, Sajal K. Das

Computer Science Faculty Research & Creative Works

Attribute graph anomaly detection aims to identify nodes that significantly deviate from the majority of normal nodes and has received increasing attention due to the ubiquity and complexity of graph-structured data in various real-world scenarios. However, current mainstream anomaly detection methods are primarily designed for centralized settings, which may pose privacy leakage risks in certain sensitive situations. Although federated graph learning offers a promising solution by enabling collaborative model training in distributed systems while preserving data privacy, a practical challenge arises as each client typically possesses a limited amount of graph data. Consequently, naively applying federated graph learning directly to …

The Age Of Synthetic Realities: Challenges And Opportunities, João Phillipe Cardenuto, Jing Yang, Rafael Padilha, Renjie Wan, Daniel Moreira, Haoliang Li, Shiqi Wang, Fernanda Andaló, Sébastien Marcel, Anderson Rocha

Computer Science: Faculty Publications and Other Works

Synthetic realities are digital creations or augmentations that are contextually generated through the use of Artificial Intelligence (AI) methods, leveraging extensive amounts of data to construct new narratives or realities, regardless of the intent to deceive. In this paper, we delve into the concept of synthetic realities and their implications for Digital Forensics and society at large within the rapidly advancing field of AI. We highlight the crucial need for the development of forensic techniques capable of identifying harmful synthetic creations and distinguishing them from reality. This is especially important in scenarios involving the creation and dissemination of fake news, …

Deciphering Trends And Tactics: Data-Driven Techniques For Forecasting Information Spread And Detecting Coordinated Campaigns In Social Media, Kin Wai Ng Lugo

USF Tampa Graduate Theses and Dissertations

The main objective of this dissertation is to develop models that predict and investigate the spread of information in social media over time. In this context, we consider topics of discussions as the information that spreads. Thus, we are interested in forecasting the number of messages per day in a future interval of time. We take a data-driven approach, in which we compare our results with real datasets from a multitude of socio-political contexts and from multiple social media platforms, specifically, Twitter and YouTube.

We identified a number of challenges related to forecasting social media time series per topic. First, …

Optimizing E-Payment Applications For Older Adults: User-Centered Solutions To Improve Security, Privacy, Usability, And Accessibility, Urvashi Kishnani

Electronic Theses and Dissertations

In an increasingly digital world, older adults are rapidly becoming a vital demographic in the realm of electronic financial transactions. It is imperative to address their unique needs and challenges to ensure their financial well-being. Older adults can be more vulnerable to various online threats, making security and privacy paramount. As they adapt to the digital age, understanding their specific privacy concerns and preferences is crucial for creating trustworthy e-payment systems. Moreover, enhancing the usability of e-payment applications for older adults promotes financial independence and inclusion, contributing to their overall quality of life. By focusing on these critical dimensions, we …

Voucher Abuse Detection With Prompt-Based Fine-Tuning On Graph Neural Networks, Zhihao Wen, Yuan Fang, Yihan Liu, Yang Guo, Shuji Hao

Research Collection School Of Computing and Information Systems

Voucher abuse detection is an important anomaly detection problem in E-commerce. While many GNN-based solutions have emerged, the supervised paradigm depends on a large quantity of labeled data. A popular alternative is to adopt self-supervised pre-training using label-free data, and further fine-tune on a downstream task with limited labels. Nevertheless, the "pre-train, fine-tune" paradigm is often plagued by the objective gap between pre-training and downstream tasks. Hence, we propose VPGNN, a prompt-based fine-tuning framework on GNNs for voucher abuse detection. We design a novel graph prompting function to reformulate the downstream task into a similar template as the pretext task …

Program Analysis For Android Security And Reliability, Sydur Rahaman

Dissertations

The recent, widespread growth and adoption of mobile devices have revolutionized the way users interact with technology. As mobile apps have become increasingly prevalent, concerns regarding their security and reliability have gained significant attention. The ever-expanding mobile app ecosystem presents unique challenges in ensuring the protection of user data and maintaining app robustness. This dissertation expands the field of program analysis with techniques and abstractions tailored explicitly to enhancing Android security and reliability. This research introduces approaches for addressing critical issues related to sensitive information leakage, device and user fingerprinting, mobile medical score calculators, as well as termination-induced data loss. …

Victimization By Deepfake In The Metaverse: Building A Practical Management Framework, Julia Stavola, Kyung-Shick Choi

International Journal of Cybersecurity Intelligence & Cybercrime

Deepfake is digitally altered media aimed to deceive online users for political favor, monetary gain, extortion, and more. Deepfakes are the prevalent issues of impersonation, privacy, and fake news that cause substantial damage to individuals, groups, and organizations. The metaverse is an emerging 3-dimensional virtual platform led by AI and blockchain technology where users freely interact with each other. The purpose of this study is to identify the use of illicit deep fakes which can potentially contribute to cybercrime victimization in the metaverse. The data will be derived from expert interviews (n=8) and online open sources to design a framework …

Increasing The Efficiency And Accuracy Of Collective Intelligence Methods For Image Classification, Md Mahmudulla Hassan

Open Access Theses & Dissertations

Collective intelligence has emerged as a powerful methodology for annotating and classifying challenging data that pose difficulties for automated classifiers. It works by leveraging the concept of "wisdom of the crowds" which approximates a ground truth after aggregating experts' feedback and filtering out noise. However, challenges arise when certain applications, such as medical image classification, security threat detection, and financial fraud detection, demand accurate and reliable data annotation. The unreliability of experts due to inconsistent expertise and competencies, coupled with the associated cost and time-consuming judgment extraction, presents additional challenges.

Input aggregation is the process of consolidating and combining multiple …

Invading The Integrity Of Deep Learning (Dl) Models Using Lsb Perturbation & Pixel Manipulation, Ashraful Tauhid

Theses and Dissertations

The use of deep learning (DL) models for solving classification and recognition-related problems are expanding at an exponential rate. However, these models are computationally expensive both in terms of time and resources. This imposes an entry barrier for low-profile businesses and scientific research projects with limited resources. Therefore, many organizations prefer to use fully outsourced trained models, cloud computing services, pre-trained models are available for download and transfer learning. This ubiquitous adoption of DL has unlocked numerous opportunities but has also brought forth potential threats to its prospects. Among the security threats, backdoor attacks and adversarial attacks have emerged as …

Generalizing Graph Neural Networks Across Graphs, Time, And Tasks, Zhihao Wen

Dissertations and Theses Collection (Open Access)

Graph-structured data are ubiquitous across numerous real-world contexts, encompassing social networks, commercial graphs, bibliographic networks, and biological systems. Delving into the analysis of these graphs can yield significant understanding pertaining to their corresponding application fields.Graph representation learning offers a potent solution to graph analytics challenges by transforming a graph into a low-dimensional space while preserving its information to the greatest extent possible. This conversion into low-dimensional vectors enables the efficient computation of subsequent graph algorithms. The majority of prior research has concentrated on deriving node representations from a single, static graph. However, numerous real-world situations demand rapid generation of representations …

Scalable And Globally Optimal Generalized L1 K-Center Clustering Via Constraint Generation In Mixed Integer Linear Programming, Aravinth Chembu, Scott Sanner, Hassan Khurram, Akshat Kumar

Research Collection School Of Computing and Information Systems

The k-center clustering algorithm, introduced over 35 years ago, is known to be robust to class imbalance prevalent in many clustering problems and has various applications such as data summarization, document clustering, and facility location determination. Unfortunately, existing k-center algorithms provide highly suboptimal solutions that can limit their practical application, reproducibility, and clustering quality. In this paper, we provide a novel scalable and globally optimal solution to a popular variant of the k-center problem known as generalized L1 k-center clustering that uses L1 distance and allows the selection of arbitrary vectors as cluster centers. We show that this clustering objective …

Scalable And Globally Optimal Generalized L1 K-Center Clustering Via Constraint Generation In Mixed Integer Linear Programming, Aravinth Chembu, Scott Sanner, Hassan Khurran, Akshat Kumar

Research Collection School Of Computing and Information Systems

The k-center clustering algorithm, introduced over 35 years ago, is known to be robust to class imbalance prevalent in many clustering problems and has various applications such as data summarization, document clustering, and facility location determination. Unfortunately, existing k-center algorithms provide highly suboptimal solutions that can limit their practical application, reproducibility, and clustering quality. In this paper, we provide a novel scalable and globally optimal solution to a popular variant of the k-center problem known as generalized L_1 k-center clustering that uses L_1 distance and allows the selection of arbitrary vectors as cluster centers. We show that this clustering objective …

Reinventing Cybersecurity Internships During The Covid-19 Pandemic, Lori L. Sussman

Journal of Cybersecurity Education, Research and Practice

The Cybersecurity Ambassador Program provides professional skills training for emerging cybersecurity professionals remotely. The goal is to reach out to underrepresented populations who may use Federal Work-Study (FWS) or grant sponsored internships to participate. Cybersecurity Ambassadors (CAs) develop skills that will serve them well as cybersecurity workers prepared to do research, lead multidisciplinary, technical teams, and educate stakeholders and community members. CAP also reinforces leadership skills so that the next generation of cybersecurity professionals becomes a sustainable source of management talent for the program and profession. The remote curriculum innovatively builds non-technical professional skills (communications, teamwork, leadership) for cybersecurity research …

A Machine Learning Approach To Deepfake Detection, Delaney Conrad

All Undergraduate Theses and Capstone Projects

The ability to manipulate videos has been around for decades but a process that once would take time, money, and professionals, can now be created by anyone due to the rapid advancement of deepfake technology. Deepfakes use deep learning artificial intelligence to make fake digital content, typically in the form of swapping a person’s face in a video or image. This technology could easily threaten and manipulate individuals, corporations, and political organizations, so it is essential to find methods for detecting deepfakes. As the technology for creating deepfakes continues to improve, these manipulated videos are becoming increasingly undetectable. It is …

A New Credit Scoring Model To Reduce Potential Predatory Lending: A Design Science Approach, Anna Zakowska

CGU Theses & Dissertations

This research examines the potential impact of implementing a novel credit scoring model that integrates attributes beyond the traditional FICO model. It aims to address issues related to predatory lending and the financial exclusion affecting individuals often categorized as 'credit invisible,' 'credit unscorable,' 'unbanked,' and 'underbanked.' These individuals typically face difficulties in establishing or repairing a credit history, which poses a challenge for financial institutions in accurately evaluating their creditworthiness. This gap in the credit assessment process often opens doors to unfair lending practices. To tackle this problem, a systematically designed, built, tested, and evaluated innovative credit scoring model was …

Application Of Big Data Technology, Text Classification, And Azure Machine Learning For Financial Risk Management Using Data Science Methodology, Oluwaseyi A. Ijogun

College of Graduate Studies: Theses & Dissertations

Data science plays a crucial role in enabling organizations to optimize data-driven opportunities within financial risk management. It involves identifying, assessing, and mitigating risks, ultimately safeguarding investments, reducing uncertainty, ensuring regulatory compliance, enhancing decision-making, and fostering long-term sustainability. This thesis explores three facets of Data Science projects: enhancing customer understanding, fraud prevention, and predictive analysis, with the goal of improving existing tools and enabling more informed decision-making. The first project examined leveraged big data technologies, such as Hadoop and Spark, to enhance financial risk management by accurately predicting loan defaulters and their repayment likelihood. In the second project, we investigated …

Role-Reversibility, Ai, And Equitable Justice — Or: Why Mercy Cannot Be Automated, Stephen E. Henderson, Kiel Brennan-Marquez

Faculty Articles

A few years ago, we developed the concept of “role-reversibility” in AI governance: the idea that it matters whether a party exercising judgment is reciprocally vulnerable to the effects of judgment. This idea, we argued, supplies a deontic reason to maintain certain spheres of human judgment even if (or when) truly intelligent machines become demonstrably superior in every utilitarian sense. While computer science remains far from that holy grail, generative AI is raging through systems as diverse as healthcare, finance, advertising, law, and academe, making it imperative to further shore up our claim. We do so by situating role-reversibility within …

A Longitudinal Study Of Factors That Affect User Interactions With Social Media And Email Spam, Wojciech M. Mazurek

Graduate Theses, Dissertations, and Problem Reports (ETD)

Given the rapid growth of social media and the increasing prevalence of spam, it is crucial to understand users’ interactions with unsolicited content to develop effective countermeasures against spam. This thesis focuses on exploring the factors that influence users’ decisions to interact with spam on social media and email. It builds upon prior work, which serves as a foundation for further research and conducting a longitudinal analysis. Our results are based on the analysis of 221 responses collected through an online survey. The survey not only gathered demographic information such as age, gender, and race but also collected data on …

Creating Data From Unstructured Text With Context Rule Assisted Machine Learning (Craml), Stephen Meisenbacher, Peter Norlander

School of Business: Faculty Publications and Other Works

Popular approaches to building data from unstructured text come with limitations, such as scalability, interpretability, replicability, and real-world applicability. These can be overcome with Context Rule Assisted Machine Learning (CRAML), a method and no-code suite of software tools that builds structured, labeled datasets which are accurate and reproducible. CRAML enables domain experts to access uncommon constructs within a document corpus in a low-resource, transparent, and flexible manner. CRAML produces document-level datasets for quantitative research and makes qualitative classification schemes scalable over large volumes of text. We demonstrate that the method is useful for bibliographic analysis, transparent analysis of proprietary data, …

Post Pandemic Cyberbiosecurity Threats From Terrorist Groups, Haley D. Dodge

Master's Theses

The research in this thesis explored the research question: Are United States (US) health systems accessible to cyber-bio terrorist attacks post-pandemic, within the context of the emerging discipline of cyberbiosecurity? Key findings of the analysis demonstrated how US health systems are more accessible to cyber-bio terrorist attacks specifically from cyber hacking groups based on the increasing sophistication of their cyber capabilities and the lack of cyber protection for biological systems. The concept of cyberbiosecurity was first introduced in 2018 by researchers exploring the converging threat landscape of the cyber and biology domains. As biology is growing more dependent upon vulnerable …

Dynamics Of Dark Web Financial Marketplaces: An Exploratory Study Of Underground Fraud And Scam Business, Bo Ra Jung, Kyung-Shick Choi, Claire Seungeun Lee

International Journal of Cybersecurity Intelligence & Cybercrime

The number of Dark Web financial marketplaces where Dark Web users and sellers actively trade illegal goods and services anonymously has been growing exponentially in recent years. The Dark Web has expanded illegal activities via selling various illicit products, from hacked credit cards to stolen crypto accounts. This study aims to delineate the characteristics of the Dark Web financial market and its scams. Data were derived from leading Dark Web financial websites, including Hidden Wiki, Onion List, and Dark Web Wiki, using Dark Web search engines. The study combines statistical analysis with thematic analysis of Dark Web content. Offering promotions …

Gr-175 - Credit Default Prediction Of Money Borrowing Companies Using Pyspark Framework, Jitendra Sai Kota, Mallika Boyapati

C-Day Computing Showcase

Credit-lending companies have resorted to the use of Machine Learning algorithms in the recent past to predict the probability of default of a customer for future credit lending purposes. Most credit companies view this as a binary classification problem of predicting whether an individual would default or not. Companies have been using models of Logistic Regression for a long time because of the explainability of the final feature set used in modeling. Explainability brings transparency to every stakeholder involved in the process. Other models like Neural Nets have achieved better accuracy scores, but the features generated by them are not …

Challenges Of Digital Privacy In Banking Organizations, Okechukwu Innocent Ogudebe

Walden Dissertations and Doctoral Studies

As the information and technology age becomes more advanced, digital privacy flaws have become more challenging. Information technology (IT) security managers, chief information security officers, and other stakeholders in banks are concerned with identity-based authentication attacks because identity-theft attacks cause data breaches. Grounded in the protection motivation theory, the purpose of this qualitative pragmatic study was to examine strategies IT security professionals working on internet banking platforms use to mitigate identity-based authentication attacks. The study participants comprised five IT security professionals currently working in the online banking industry from the northeastern United States with at least 5 years of experience …

Few-Shot Malware Detection Using A Novel Adversarial Reprogramming Model, Ekula Praveen Kumar

Browse all Theses and Dissertations

The increasing sophistication of malware has made detecting and defending against new strains a major challenge for cybersecurity. One promising approach to this problem is using machine learning techniques that extract representative features and train classification models to detect malware in an early stage. However, training such machine learning-based malware detection models represents a significant challenge that requires a large number of high-quality labeled data samples while it is very costly to obtain them in real-world scenarios. In other words, training machine learning models for malware detection requires the capability to learn from only a few labeled examples. To address …

Camouflaged Poisoning Attack On Graph Neural Networks, Chao Jiang, Yi He, Richard Chapman, Hongyi Wu

Computer Science Faculty Publications

Graph neural networks (GNNs) have enabled the automation of many web applications that entail node classification on graphs, such as scam detection in social media and event prediction in service networks. Nevertheless, recent studies revealed that the GNNs are vulnerable to adversarial attacks, where feeding GNNs with poisoned data at training time can lead them to yield catastrophically devastative test accuracy. This finding heats up the frontier of attacks and defenses against GNNs. However, the prior studies mainly posit that the adversaries can enjoy free access to manipulate the original graph, while obtaining such access could be too costly in …

The Application Of Deep Learning And Cloud Technologies To Data Science, Ian A. Trawinski

College of Graduate Studies: Theses & Dissertations

Machine Learning and Cloud Computing have become a staple to businesses and educational institutions over the recent years. The two forefronts of big data solutions have garnered technology giants to race for the superior implementation of both Machine Learning and Cloud Computing. The objective of this thesis is to test and utilize AWS SageMaker in three different applications: time-series forecasting with sentiment analysis, automated Machine Learning (AutoML), and finally anomaly detection. The first study covered is a sentiment-based LSTM for stock price prediction. The LSTM was created with two methods, the first being SQL Server Data Tools, and the second …

Exposing And Fixing Causes Of Inconsistency And Nondeterminism In Clustering Implementations, Xin Yin

Dissertations

Cluster analysis aka Clustering is used in myriad applications, including high-stakes domains, by millions of users. Clustering users should be able to assume that clustering implementations are correct, reliable, and for a given algorithm, interchangeable. Based on observations in a wide-range of real-world clustering implementations, this dissertation challenges the aforementioned assumptions.

This dissertation introduces an approach named SmokeOut that uses differential clustering to show that clustering implementations suffer from nondeterminism and inconsistency: on a given input dataset and using a given clustering algorithm, clustering outcomes and accuracy vary widely between (1) successive runs of the same toolkit, i.e., nondeterminism, and …

Don't Bite The Bait: Phishing Attack For Internet Banking (E-Banking), Ilker Kara

Journal of Digital Forensics, Security and Law

Phishing attacks are based on obtaining desired information from users quickly and easily with the help of misdirecting, panicking, curiosity, or excitement. Most of the phishing web sites are designed on internet banking(e-banking) and the attackers can acquire financial information of misled users with the tactics and discourses they develop. Despite the increase of prevention techniques against phishing attacks day by day, an effective solution could not be found for this issue due to the human factor. Because of this reason, real phishing attack studies are essential to study and analyze the attackers’ attack techniques and strategies. This study focused …

Taxthemis: Interactive Mining And Exploration Of Suspicious Tax Evasion Group, Yating Lin, Kamkwai Wong, Yong Wang, Rong Zhang, Bo Dong, Huamin Qu, Qinghua Zheng

Research Collection School Of Computing and Information Systems

Tax evasion is a serious economic problem for many countries, as it can undermine the government’s tax system and lead to an unfair business competition environment. Recent research has applied data analytics techniques to analyze and detect tax evasion behaviors of individual taxpayers. However, they have failed to support the analysis and exploration of the related party transaction tax evasion (RPTTE) behaviors (e.g., transfer pricing), where a group of taxpayers is involved. In this paper, we present TaxThemis, an interactive visual analytics system to help tax officers mine and explore suspicious tax evasion groups through analyzing heterogeneous tax-related data. A …