Research Interests

My research focuses on the intersection of machine learning, temporal data, video learning, and representation learning. I develop advanced models to improve video understanding, particularly in compressed domains and through self-supervised techniques. Additionally, I explore novel methods for representing low-dimensional sequential data, such as protein sequences and univariate time series, using de Bruijn graphs to enhance model performance in classification and forecasting tasks.

Education

Fall 2023 – present
Ph.D. Computer Science, Indiana University, Luddy School of Informatics, Computing, and Engineering
  • Advisor: Prof. Dr. Mehmet M Dalkilic
  • Co-Advisor: Dr. Hasan Kurban
2017 – 2021
BS Computer Science, TOBB University of Economics and Technology

Research and Work Experience

Research Experience

May 2024 – Jul. 2024
Temporary Research Associate – Texas A&M University at Qatar
  • Developed a self-supervised learning framework for video data, enabling models to learn meaningful representations without labeled data.
  • Implemented federated video learning in the compressed domain, optimizing performance while preserving privacy.
Aug. 2023 – Ongoing
Student Researcher – Kurban Intelligence Labs (Visit Lab Website)
  • Conducted research and authored academic papers on time series forecasting, Type 1 Diabetes hypoglycemia detection, and protein classification.
  • Currently conducting research on video learning, self-supervised learning, and representation learning using de Bruijn graphs.

Work Experience

Jul. 2021 – Apr. 2023
Full-Stack Software Developer – Innova It Solutions (Visit Project Website)
  • Contributed to the development of the "Centralized Fault Management System (MARS)," designed to provide end-to-end fault detection, diagnosis, and resolution for telecommunication networks and IT infrastructures.
  • Responsible for improving the legacy codebase and developing new functionalities based on functional specifications and business requirements.
  • Gained experience working in an agile development environment.
  • Developed microservices using Spring Boot, interacting with PL/SQL and MongoDB databases in a microservice architecture.
  • Worked on front-to-backend interactions using React.js and Vaadin frameworks, utilizing RESTful services for seamless integration.
Spring 2023 – Ongoing
Associate Instructor (TA) – CSCI-C 200 Introduction to Computers and Programming
  • Instructor: Prof. Dr. Mehmet M Dalkilic

Awards and Recognition

2023
Fall 2023 Luddy Doctoral Associate Instructor Fellowship, Luddy School of Informatics, Computing, and Engineering.

Publications

2024
Mert Onur Cakiroglu, Hasan Kurban, Parichit Sharma, M. Oguzhan Kulekci, Elham Khorasani Buxton, Maryam Raeeszadeh-Sarmazdeh, Mehmet Dalkilic. "An Extended De Bruijn Graph for Feature Engineering Over Biological Sequential Data." Machine Learning: Science and Technology. (Impact Factor: 6.8) (DOI)
2024
Mert Onur Cakiroglu, Hasan Kurban, Elham Khorasani Buxton, Mehmet Dalkilic. "A Novel Discrete Time Series Representation with De Bruijn Graphs for Enhanced Forecasting Using TimesNet (Extended Abstract)." 2024 IEEE 11th International Conference on Data Science and Advanced Analytics (DSAA), San Diego, CA, USA, 2024, pp. 1-3. (DOI)
2024
Mert Onur Cakiroglu, Hasan Kurban, Elham Khorasani Buxton, Mehmet Dalkilic. "A reinforcement learning approach to effective forecasting of pediatric hypoglycemia in diabetes I patients using an extended de Bruijn graph." Nature – Scientific Reports Sci Rep 14, 31251. (Impact Factor: 3.8) (DOI)
2024
Mert Onur Cakiroglu, Hasan Kurban, Elham Khorasani Buxton, Mehmet Dalkilic. "A Novel Discrete Time Series Representation with De Bruijn Graphs for Enhanced Forecasting Using TimesNet." Machine Learning Journal (Under Review).

Curriculum Vitae

If the PDF does not display, you can download the CV here.