Design Project Articles Master Degree
Permanent URI for this communityhttps://hdl.handle.net/20.500.12475/48
Como parte de las Colecciones Especiales, la Biblioteca cuenta con los Artículos de Proyecto y Carteles de Investigación (Poster) sometidos para la obtención del grado de los estudiantes de la Escuela Graduada.
Recursos disponibles:
1. Artículos de Proyectos - 2008 en adelante.
2. Carteles de Investigación - 2018 en adelante.
Browse
91 results
Search Results
Item Quantum-Safe Security for 5G Networks(Polytechnic University of Puerto Rico, 2024) González Maldonado, Cristian; Duffany, JeffreyThis research investigates post-quantum cryptography's implementation and performance impact in 5G networks using the free5GC platform. Developed a comprehensive framework incorporating CRYSTALS-Kyber algorithms through Docker-containerized network functions and OpenSSL integration. Testing with two virtualized servers (2 vCPU, 4 GB RAM) demonstrated that quantum-resistant security can be achieved without compromising network performance requirements. Results show Kyber variants maintain latency well below 1ms (Kyber512: 0.059702s, Kyber1024: 0.051873s), with manageable bandwidth requirements (Kyber512: 682.99 bytes/packet, Kyber1024: 775.79 bytes/packet). Kyber1024 achieved superior throughput (14,958.19 bytes/second) compared to traditional X25519 (10,496.87 bytes/second. A practical implementation framework through Docker containers with custom Python scripts for performance analysis enables organizations to merge quantum-resistant security with operational network efficiency. This research has developed structures that secure telecommunications frameworks from threats caused by advanced quantum computing techniques. Key Terms ⎯ 5G Network Security, CRYSTALS-Kyber, Post-Quantum Cryptography, Quantum-Safe Architecture.Item Analysis of Sentiment in YouTube Comments for 2024 Political Campaigns for Governor of Puerto Rico(Polytechnic University of Puerto Rico, 2024) Oriol Rivera, Carlos A.; Duffany, JeffreyThe analysis of public sentiment is an invaluable tool in political campaigns, enabling candidates and their teams to assess voter attitudes and adjust campaign strategies accordingly. This project focuses on collecting and analyzing video comments on YouTube from the months leading up to an election to determine the public’s sentiment toward different candidates for governor in Puerto Rico. To make this sentiment analysis, the Text Blob Python library was used due to its ease of use and robust functionality. The project incorporates both client-side and server-side components, integrating modern web development frameworks and libraries to provide good user experience. This investigation outlines the data collection methodology, system architecture, analysis process, discussion of results, and coding logic used to achieve the analysis. Key Terms ⎯ Political Campaigns, Sentiment Analysis, TextBlob Python Library.Item Artificial Intelligence-Driven Website Composer(Polytechnic University of Puerto Rico, 2024) Cepeda Delgado, Sergio R.; Duffany, JeffreyThis study outlines the creation of an artificial intelligence (AI)-powered web composer that utilizes OpenAI's Application Programming Interface (API) models GPT-4o and DALL-E-2, together with Python and HTML code, to enable the automatic creation of web pages. This composer enables users to define their design preferences and content requirements, which are subsequently translated and implemented by the AI using sophisticated natural language processing and deep learning methods. The incorporation of these technologies eliminates the need for substantial technical expertise, therefore expanding the availability of web design to a wider range of users. Comparative assessments using traditional web development approaches demonstrate a substantial improvement in efficiency and user happiness, accompanied by notable decreases in production time. This research highlights the significant impact that AI may have on web development and lays the groundwork for future investigations into AI-powered design tools. Key Terms ¾ deep learning, natural-language understanding (NLU), OpenAI API integration, user-centric design automation.Item Shodan Search Engine(Polytechnic University of Puerto Rico, 2024) Nevárez Román, Paula A.; Duffany, JeffreySearch engines have become a crucial component of the Internet. Shodan is a unique type of search engine. Unlike conventional search engines that index the World Wide Web, Shodan offers users the ability to explore all publicly accessible devices on the Internet. Shodan explores the realm of IoT devices, servers, webcams, and more, providing valuable insights into the digital landscape. This paper delves into Shodan's functionalities, including IP scanning and other services, highlighting its applications in network security, market research, and cyber risk assessment. However, alongside its benefits, Shodan also poses significant risks due to its accessibility to anyone with internet access and basic search query knowledge. This paper emphasizes the responsible and ethical use of Shodan to prevent security vulnerabilities and privacy breaches, providing a comprehensive analysis of its features, implications, and ethical considerations in today's interconnected digital landscape. Key Terms ⎯ features, risks, search engine, and Shodan.Item Design and Testing of a Sentiment Analysis Model for Digital Data(Polytechnic University of Puerto Rico, 2024) Feliz Berrios, Marcos; Duffany, JeffreyThis study analyzed sentiments on social media regarding American airlines during COVID 19 and IMDb reviews of "Tenet" using the BERT model for sentiment analysis. Results showed Delta Airlines received more positive sentiments than American and United Airlines, which faced consistent negative sentiments. Similarly, "Tenet" reviews exhibited a mix of positive and negative opinions. The BERT model demonstrated high accuracy in sentiment prediction, capturing nuanced opinions. Limitations included potential dataset bias and challenges in encompassing all factors influencing sentiment. Nonetheless, the study highlighted sentiment analysis' potential in understanding evolving public perceptions on social media and beyond. Key Terms − Data Analysis, Public Opinion, Sentiment Analysis, Social Media.Item Applications of Fast.ai Pretrained Models in Image Classification Problem(Polytechnic University of Puerto Rico, 2024) Tapia Maldonado, Antonio Ahmed; Duffany, JeffreySince the inception of Transformers and GPTs, artificial intelligence has proliferated. Fast.ai is among the cutting-edge libraries that are leading new advancements in the field. To harness the power of fast.ai and other advancements in the field, we set out to try and evaluate the practicality of the fast.ai library. To achieve this, we choose a given use case for artificial intelligence and then set out to fulfill said use case by leveraging fast.ai. We created three image classification models through fast.ai and then made an application that used those models. The use case we chose was a local wildlife fauna and flora classifier. The results from training the models were models with meager error rates, and these models had little to no data engineering. Key Terms ¾ computer vision, deep learning, fast.ai, image classification.Item Cybersecurity Tools(Polytechnic University of Puerto Rico, 2024) López Ubinas, Harry E.; Duffany, JeffreyThe goal of this project is to analyze the effectiveness of different Cybersecurity tools. It will test different techniques such as vulnerability scanning on the network, Wi-Fi hacking, and password cracking. Key Terms ¾ cybersecurity, password cracking, vulnerability scanning, wi-fi-hacking.Item Employee Guide to Cybersecurity and Cyber Threats(Polytechnic University of Puerto Rico, 2023) Feliciano Cruz, Emily Y.; Duffany, JeffreyCybersecurity is a growing concern for companies and workers due to the growing threats posed by technology. The Covid-19 pandemic has highlighted the need for protective shields, with malware being a major threat. WannaCry, a $4 billion attack, cost businesses up to $4 billion [1]. While technical solutions like spam filters protect users, human error remains a risk factor that needs to be addressed to improve security across various systems and communications. EFC Secure is committed to a modern approach to cybersecurity to prevent unauthorized access, disclosure, and damage to electronic information and critical infrastructure. This manual provides guidance on employees' roles and responsibilities to help prevent incidents that put at risk the information handled by EFC Secure and its customers. By focusing on information security, EFC Secure ensures effective controls for its customers and employees. Key Terms - Antivirus, Incident, Phishing, Ransomware.Item ZAP Proxy and OWASP Top 10(Polytechnic University of Puerto Rico, 2023) Ramos Flores, Eduard; Duffany, JeffreyThe Zed Attack Proxy is a well-known and popular assessment tool in the cybersecurity community. The Open Web Application Security Project community offers, develops, and maintains the Zed Attack Proxy. The Open Web Application Security Project community also publishes the top ten security risks faced by web applications. Paired with the Zed Attack Proxy, The Open Web Application Security Project’s top 10 security risks publication, serves as a baseline for security professionals assessing the security compliance of web applications. This study aims to evaluate the effectiveness and efficiency of the Open Web Application Security Project’s Zed Attack Proxy tool against real-world production web applications and vulnerable by design penetration labs web applications. Key Terms ⎯ Open-Source, Penetration Testing, Security Assessment, Web Applications.Item Network Security Assessment using Bettercap: DNS Spoofing and How to Mitigate(Polytechnic University of Puerto Rico, 2023) Santiago Solivan, Rafael; Duffany, JeffreyIn the ever-evolving landscape of cybersecurity, the demand for effective tools and techniques to safeguard networks from malicious actors is paramount. This master's degree final project proposes an investigation into the capabilities of the Bettercap tool for comprehensive network vulnerability assessment and DNS spoofing attack simulation. The project's novelty lies in its utilization of a virtual lab environment to create a controlled, simulated network, enabling the practical demonstration of vulnerability identification and potential risks associated with DNS spoofing. Through this project, I aim to contribute to the advancement of cybersecurity knowledge and practices. Key Terms ⎯ Bettercap, DNS Spoofing, Network Vulnerability, Virtual Lab.