Overview
The midterm capstone integrates Weeks 1 to 7: research and development concepts, research methods, literature analysis, funding programmes and proposal writing [1, 2, 3, 4]. Each student writes a complete research proposal on a topic in computer science, using either the TÜBİTAK or the NSF structure.
Formats
The midterm is a complete research proposal on a computer science topic, written in English with the structure of one of two templates. With the TÜBİTAK structure, the proposal follows the current 1001 application form, or the 1002-A form for a shorter project: summaries in Turkish and English of up to 600 words each, a main text of no more than 25 pages excluding annexes for 1001, and sections on originality, method, project management with a work plan and a risk plan with alternatives, and wider impact [5, 6]. With the NSF structure, the proposal contains a one-page Project Summary with an overview and separate statements on intellectual merit and broader impacts, and a Project Description that is usually limited to 15 pages [7]. In both cases the current official template is authoritative. The weights below are indicative; in active semesters the instructor confirms them at the start of the semester.
Suggested research topics
The titles below are starting points. Students may narrow a title or propose a related one that fits the scope of the capstone.
- Privacy-preserving federated learning for clinical decision support across hospitals
- Energy-aware scheduling of AI training workloads in university data centres
- Explainable anomaly detection for industrial control systems in small and medium enterprises
- Speech recognition for Turkish public services with a fairness evaluation across dialects
- Formally verified smart contracts for transparent public procurement
- A digital twin of a campus energy system for forecasting and optimisation
- Robustness certification of vision models for autonomous agricultural machinery
- Automated feedback and program repair for introductory programming courses
- Edge computing pipelines for earthquake early warning
- Trustworthy retrieval-augmented generation for Turkish legal documents
Report
The submission contains the proposal in the chosen structure, a budget with a justification that links every item to tasks, a Gantt chart, a risk register and a one-page cover note that names the target programme and explains how each evaluation criterion is addressed. The proposal may build on the weekly tasks of Weeks 1 to 7. Sources are cited in square brackets.
A common structure for reports is given in the report template.
Assessment criteria
| Criterion | Weight | What is assessed |
|---|---|---|
| Originality and scientific quality | 25% | A clearly defined gap, verifiable objectives and a convincing position against the literature. |
| Method and feasibility | 25% | Methods that reach every objective, with adequate design, data and evaluation. |
| Project management | 20% | A realistic work plan, risks with alternatives and a budget linked to tasks. |
| Wider impact | 15% | Named users, pathways to impact and measures of progress. |
| Compliance and clarity | 15% | Adherence to the chosen template, page limits and clear writing. |
Submission
This capstone also supports self-learning and can be completed at any pace. When the course is taught actively in a semester, the midterm capstone is sent by e-mail to utkukose@sdu.edu.tr or utkukose@gmail.com no later than 23:53 (Türkiye time) on the last Sunday of Week 7, with the report and all code files attached or linked.
References
[1] OECD (2015). Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development. OECD Publishing. https://doi.org/10.1787/9789264239012-en
[2] Stol, K.-J., & Fitzgerald, B. (2018). The ABC of software engineering research. ACM Transactions on Software Engineering and Methodology, 27(3), 11. https://doi.org/10.1145/3241743
[3] Kitchenham, B., & Charters, S. (2007). Guidelines for performing systematic literature reviews in software engineering. Keele University and Durham University. EBSE Technical Report EBSE-2007-01.
[4] Doran, G. T. (1981). There's a S.M.A.R.T. way to write management's goals and objectives. Management Review, 70(11), 35-36.
[5] TÜBİTAK (2026). 1001 Bilimsel ve Teknolojik Araştırma Projelerini Destekleme Programı: Proje başvuru formu. https://tubitak.gov.tr/sites/default/files/20689/1001_basvuru_formu.doc
[6] TÜBİTAK (2026). 1002 Hızlı Destek Programı: 1002-A proje başvuru formu. https://tubitak.gov.tr/sites/default/files/20689/1002_a_basvuru_formu.doc
[7] National Science Foundation (2026). Merit review: Intellectual merit and broader impacts. https://www.nsf.gov