Supervision is open at every level in the Department of Electrical and Electronics Engineering at Kırıkkale University (KKÜ): undergraduate projects, thesis MSc and PhD. Here is what the work looks like, how to get in touch, and how admission and funding work.
Türkçe: Öğrenciler sayfası · 2026–27 lectures: Probability Theory & Neural Signals (details to follow) →
Whichever level you are at, the shape of the work is the same: a clear question, a formal method, a real case to test it on, and code that someone else can re-run.
Undergraduate
A graduation design project (bitirme projesi) for EEM students, taken from one of the directions below and scoped so it can be finished within the project period.
A project can also become a TÜBİTAK application: 2209-A for university students’ research projects, or 2209-B for industry-oriented projects run with a company. You are the applicant; I act as your academic advisor.
Master’s
A thesis MSc in the EEM programme of KKÜ’s Graduate School of Natural and Applied Sciences (Fen Bilimleri Enstitüsü). One well-posed question, a formal method, a real dataset or model to test it on, open code, and a paper as the aim.
Admission goes through the Graduate School’s call: see §05.
Doctoral
A multi-year research line in one of the three threads, with papers along the way. It can build on the published MOSOF and NDCI methods or open a new direction in calibrated uncertainty or low-cost sensing.
Co-supervision with another university is possible case by case. When a company partner is involved, the industrial PhD programme (TÜBİTAK 2244) can be a route.
These are starting points, not funded positions. Each can be scoped down to a graduation project or up to a doctoral line, and the final question is one we shape together. All six come back to the central question of my research agenda: how do we make better decisions when signals are incomplete, noisy and expensive?
Which sensors should a system carry, and where? Treat the choice as a search over competing objectives such as cost, diagnostic coverage and reliability, using evolutionary algorithms like NSGA-II/III or MOEA/D. Cases can come from aircraft subsystems, industrial plant or power-distribution networks.
NSGA-II · NSGA-III · MOEA/D · Pareto fronts
NDCI scores each sensor by how well it separates faults, how sensitive it is to fault severity, and what it adds that no other sensor does. Take it beyond the thesis models onto open fault datasets: where does the ranking hold, and where does it break, for example when degradation shifts over time?
NDCI · sensor ranking · mRMR baseline
A commodity microphone can carry a surprising amount of diagnostic information. Build and test methods that flag abnormal sound from rotating machines such as motors and pumps, and measure honestly how far a cheap sensor can go before it needs help.
spectral features · anomaly detection · baselines
A diagnosis that returns a short set of plausible faults with a stated coverage is often more useful than one confident label. Use conformal prediction and related methods to attach calibrated uncertainty to fault classifiers and remaining-useful-life estimates.
conformal prediction · PHM · remaining useful life
Which open-weight language model is worth running on an ordinary laptop with no GPU, and how good is it in Turkish? Measure speed, memory and answer quality on a reproducible benchmark, and make every result something another person can verify.
benchmarking · quantised models · Turkish evaluation
Compare multi-objective algorithms fairly on standard test problems with quality indicators such as hypervolume, and work on making Pareto fronts readable for the people who have to choose one point on them.
test suites · hypervolume · decision support
Your own idea is welcome if it connects to one of the three threads. Send a paragraph about it with your enquiry.
No prior research experience is needed.
Students rarely struggle because the maths is hard; they struggle because the framing is unfamiliar.
Student enquiry: <level>, for example Student enquiry: MSc.I read every message.
Graduate admission itself goes through the Graduate School’s call (§05). Getting in touch with a potential supervisor before you apply is encouraged: it is the best way to find out whether a direction suits you.
Your enquiry, in one email
Open an email with the subject filled in
Supervision and admission are separate steps. We can talk about a direction at any time. Admission to the MSc and PhD goes through Kırıkkale University’s Graduate School of Natural and Applied Sciences (Fen Bilimleri Enstitüsü), which opens calls each semester.
EEM runs a thesis MSc and a PhD. Applications are made online through the university’s student information system (OBS) during the call period.
| Requirement | MSc (thesis) | PhD |
|---|---|---|
| ALES | At least 55 (SAY score type) | At least 60 |
| Prior degree | Undergraduate degree; your GPA counts towards the score | A thesis MSc, with an MSc GPA of at least 3.00 / 4 |
| Language | See the call | YDS or YÖKDİL of at least 55, or an accepted equivalent |
| Subject exam | Written subject exam (bilim sınavı) | Written subject exam (bilim sınavı) |
Typical requirements, from the autumn 2026 call. The official call is binding.
These are the programmes that fit the levels on this page. Who applies differs: some you apply for yourself, others are applied for by your advisor, the university or a company partner.
Üniversite Öğrencileri Araştırma Projeleri Destekleme Programı
You apply with your own project, alone or as a small team. Your academic advisor approves it in TÜBİTAK’s system. Projects run for up to 12 months.
Üniversite Öğrencileri Sanayiye Yönelik Araştırma Projeleri Desteği Programı
A project run with a company. It needs an academic advisor and an industry advisor who works at the company.
Yurt İçi Yüksek Lisans Burs Programları
Scholarships for thesis MSc students, who apply themselves. 2210-D is the industry-oriented track.
Yurt İçi Doktora Burs Programları
Scholarships for doctoral students. You apply yourself, through TÜBİTAK’s e-BİDEB system.
Sanayi Doktora Programı
Doctoral work carried out with a company. The university and the company apply together.
Bilimsel Araştırma Projeleri
Kırıkkale University’s internal research projects, including graduate-thesis projects led by the thesis advisor.
When a funded research project has a place for a student, it will be announced on this page.
TÜBİTAK calls usually open in spring and autumn. Amounts, eligibility rules and deadlines change from call to call, so they are not repeated here: always check the current call text on tubitak.gov.tr (opens in a new tab).
No. Research is learned by doing it. What helps is comfort with the basic maths, some programming and genuine curiosity (see §03). We start with a small, well-defined first step and build from there.
Yes, and it often makes for the best projects, as long as the company can share enough data or access for the work to be real. Possible routes include a graduation project, a TÜBİTAK 2209-B project with an industry advisor, an MSc thesis, or a 2244 industrial PhD.
Work with companies runs through university channels, with data, IP and publication terms agreed before the work starts. Companies can read how that works on the industry collaboration page.
You need to be able to read English, because most of the literature is in English. Our meetings can be in Turkish or English. For PhD admission, the Graduate School asks for a YDS or YÖKDİL score (or an accepted equivalent); the call for your semester sets the details, including any requirement for the MSc.
Co-supervision, for example as a second supervisor (ikinci danışman), is possible case by case, when the topic fits and both institutions’ rules allow it. It helps if your main supervisor knows you are getting in touch.
Not automatically. TÜBİTAK scholarships such as 2210 and 2211 are applications you make yourself, and 2209-A/B are project grants you apply for with an advisor. Any position funded by a research project will be announced on this page. See funding above.
No. As your supervisor I guide the question, the method and the structure, and I give detailed feedback on the drafts you write. The writing stays yours. Writing theses, papers or projects for others is prohibited under Law 2547 (Ek md. 47), and it would also take away the part of research that is yours to learn.
Yes, if it connects to one of the three threads: sensor-suite design as a multi-objective decision, diagnostics with honest uncertainty, or low-cost sensing and efficient AI. Describe it in a paragraph in your enquiry email.
I read every message. If there is a likely fit, the next step is a conversation about a direction and what a first piece of work could look like.