About

Umit Karabiyik, PhD, is an Associate Professor in the Department of Computer and Information Technology at Purdue University. He serves as the Director of the Ubiquitous and Mobile Investigative Techniques and Technologies Lab. His research interests broadly encompass Digital Forensics, Cybersecurity, Forensic Intelligence, User and Data Privacy, Artificial Intelligence in Security, Privacy, and Forensic Applications. Dr. Karabiyik has secured federal and industrial funding from organizations such as the U.S. National Institute of Justice, U.S. Department of Homeland Security, U.S. Federal Emergency Management Agency, U.S. Bureau of Justice Assistance, The National Air and Space Intelligence Center (NASIC), and Lockheed Martin Corporation. He has developed and delivered numerous mobile and IoT forensics training courses and provided technical assistance for law enforcement and justice system professionals. Prior to his current appointment, he was an Assistant Professor in the Department of Computer Science at Sam Houston State University from 2015 to 2018. Dr. Karabiyik holds a B.S. in Computer Systems Teaching from Sakarya University, obtained in 2006, and both M.S. and Ph.D. degrees in Computer Science from Florida State University, earned in 2010 and 2015 respectively. He is actively involved in scholarly publishing, editorial roles, and conference organization within his field.

Research topics

• Computer Science
• Computer Security
• Artificial Intelligence
• Data science
• Risk analysis (engineering)
• Data Mining
• World Wide Web
• Engineering
• Software engineering
• Business
• Medicine
• Geography
• Internet privacy
• Operating system
• Cartography

Selected publications

• Mobile Forensic and Security Analysis of RedNote: A Cross-Platform Investigation on Android and iOS

  SSRN Electronic Journal · 2026-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Forensic Analysis and Privacy Implications of Llm Apps: A Case Study of Chatgpt, Copilot, and Gemini

  SSRN Electronic Journal · 2025-01-01 · 2 citations

  preprintOpen accessSenior author
  PublisherDOI

• Forensic analysis and privacy implications of LLM mobile apps: A case study of ChatGPT, Copilot, and Gemini

  Forensic Science International Digital Investigation · 2025-08-08

  articleOpen accessSenior author

  Large-Language Model (LLM) apps allow users to ask questions and receive customized responses based on their prompted inputs. They possess deep language comprehension, human-like text generation capabilities, contextual awareness, and robust problem-solving skills, making them invaluable in various domains (e.g., search engines, customer support, translation). Due to the popularity and easy access of these apps, people tend to use them to help them in their daily tasks. However, when downloaded on mobile phones, these apps access various types of personal data (e.g., account information, location, and device information), including browser cookies, which may be of forensic value. The permissions that these applications request can be abused by developers to over-collect data about their users. Therefore, this paper aims to analyze the data these apps (ChatGPT, Gemini, and Copilot) access and find valuable forensic artifacts, including those that may cause privacy concerns and potential risks and threats associated with their use. It also analyzes the type of data stored when downloaded and accessed through a mobile device. The results show that ChatGPT and Copilot stores conversation data in plain text, along with browser data, on both Android and iOS. Gemini, on the other hand, stores all conversation data, browser data, and images in the cloud, which can be extracted through Google Takeout. • Forensic and privacy analysis of ChatGPT, Gemini, and Co-pilot mobile applications on Android and iOS, as well as the cloud for Gemini. • Chatbot conversations were recovered for ChatGPT, Gemini, and Co-pilot, depending on the platform they are used on. • Location data was accessible within a radius of .5 miles, even with location settings turned off on an iOS device for Gemini and ChatGPT. • Details, including user prompts, browser data, and location data, were recovered from Copilot on an Android device.

  PublisherDOI

• Finding What Keywords Miss: Vector Search for Digital Forensics

  2025-01-01

  articleOpen accessSenior author

  Digital investigators face challenges searching through massive data when coded language evades keyword searches. Our research shows how vector embeddings enhance digital forensics by detecting implicit drug references in social media. Using LLM2Vec with instruction-based embeddings, we significantly outperformed traditional keyword searches when dealing with slang, enabling investigators to find content based on meaning rather than exact word matching.

  PublisherOA PDFDOI

• From Screen to Scene: An In-Depth Forensic Analysis on Deepseek

  SSRN Electronic Journal · 2025-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Forensic examination of tencent QQ mobile application features for evidence collection on android and iOS devices

  Discover Computing · 2025-10-25

  articleOpen accessSenior author

  Abstract Tencent QQ, established in 1999, ranks among the most extensively utilized instant messaging applications globally. At its peak, it attracted approximately 900 million users (Business Daily in Sohu 2023). Despite the emergence of numerous similar applications such as WeChat, QQ continues to hold a strong position within business and interest-based communities, particularly appealing to young adults. Forensic examinations of QQ have been ongoing since 2009, primarily addressing its memory function, instant messaging capabilities, and desktop version. However, there is a notable lack of thorough research on the many beneficial functions of the mobile version of QQ on both Android and iOS platforms. Additionally, online fraudsters, especially scam groups in Southeast Asia, have taken advantage of the app. Inadvertently, QQ provides scammers with the means to contact victims and extort property. To bridge this knowledge gap, this study performs a forensic analysis of QQ’s new features on Android and iOS. Our study covers new capabilities such as device detection, file editing and transferring, the integrated camera, document sharing, payment, and service functions, message withdrawal, real-time location sharing, phone numbers, contacts, QQ group activities, chat history with its backup, QQ zone, and privacy preservation and protection measures. The objective of this research is to assist investigators by enhancing the use of forensic tools concerning QQ and its new functionalities to discern what evidence can be acquired and recovered, thereby closing the existing knowledge gap.

  PublisherOA PDFDOI

• Evaluating Sustainability Trade-Offs in Engineering Design as System of Systems Using LLMs

  2025-06-08

  articleSenior author

  Design thinking is a critical skill in 21st-century engineering education, fostering creative problem-solving and user-centered design approaches. When integrated with systems thinking, particularly the Distinctions, Systems, Relationships, and Perspectives (DSRP) framework, it equips engineers with a structured approach to tackling complex challenges. Sustainability, encompassing environmental, economic, and social dimensions, represents a crucial intersection of these frameworks. Despite the increasing incorporation of sustainability in engineering curricula, there is limited research on how students navigate trade-offs among these dimensions in practical applications. This study explores how engineering students internalize and apply sustainability principles in an energy-efficient housing design challenge. Using Energy3D software, students designed energy-efficient houses while balancing cost-effectiveness, environmental impact, and technical feasibility. We analyzed student reflections and design outcomes using OpenAI's ChatGPT-4.0 model to identify sustainability themes and assess trade-offs. Our findings reveal that while most students effectively balanced environmental and economic considerations, variations in prioritization led to distinct design strategies. The majority (52.4%) were classified as Balanced Sustainability Champions, excelling in both dimensions, while others emphasized either technological innovation or cost constraints. The results highlight disparities in environmental performance, suggesting opportunities for targeted instructional improvements. By leveraging large language models (LLMs) for thematic analysis, this study provides novel insights into sustainability education, demonstrating how integrating design thinking and systems thinking within experiential learning can enhance students' ability to address real-world sustainability challenges.

  PublisherDOI

• The Hidden Realms of Router Apps: Forensic Analysis of TP-Link Tether and ASUS Router

  Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · 2025-01-01 · 1 citations

  book-chapterSenior author
  PublisherDOI

• From TikTok to RedNote: A Comprehensive Mobile Forensic and Security Analysis on Android

  2025-04-24

  articleSenior author

  RedNote, known in China as Xiaohongshu, has gained substantial traction in the US market following Tik-Tok's recent restrictions. To address the forensics gap of this app, this paper presents a comprehensive forensic analysis of the latest RedN ote version on an Android device. The study involves populating a Google Pixel 5a with typical user data, followed by the image acquisition and detailed examination using Magnet AXIOM, and validation with ALEAPP. In parallel, a static analysis of the RedN ote APK was conducted using the Mobile Security Framework (MobSF) to identify potentially risky permissions, cryptographic weaknesses, and insecure configurations. Our research identifies and analyzes various types of user data, including account information, interaction records, content browsing and posting. We also delve into the app's folder structure and data storage paths, highlighting the forensic value of this recently popular application, which poses unique challenges due to its non-international development and singular APK release across regions.

  PublisherDOI

• Forensic Analysis of Tencent QQ: Investigating New Mobile Features for Evidence Collection on Android and iOS Devices

  Research Square · 2025-07-07

  preprintOpen accessSenior author
  PublisherOA PDFDOI

Frequent coauthors

• Shinelle Hutchinson

  Purdue University West Lafayette

  26 shared

• Yung Han Yoon

  Purdue University West Lafayette

  20 shared

• Neesha Shantaram

  Purdue University West Lafayette

  17 shared

• Mohammad Meraj Mirza

  Taif University

  12 shared

• Fahad E. Salamh

  Purdue University System

  10 shared

• Sudhir Aggarwal

  Florida State University

  8 shared

• Marcus Rogers

  Purdue University System

  8 shared

• Miloš Stanković

  Purdue University West Lafayette

  4 shared

Labs

• UMIT2 LabPI

  Current & Past Students

Education

• PhD, Computer Science

  Florida State University

  2015

• MS, Computer Science

  Florida State University

  2010

• BS, Computer Systems Teaching

  Sakarya Üniversitesi

  2006

Awards & honors

• Department of Computer and Information Technology, Outstandi…
• Department of Computer and Information Technology, Outstandi…
• People’s Choice Poster Award Winner, 23rd Annual Emergency M…
• The Seed for Success Award, Excellence in Research, Purdue U…
• Outstanding ACM Chapter Advisor, Sam Houston State Universit…