Junior Data Scientist at Muğla Sıtkı Koçman University · MSc Statistics (in progress) · Book author · Bioconductor contributor
I build things that turn messy institutional data into decisions people actually use — data dictionaries, SQL models, and R Shiny applications that go into production rather than into a folder.
- Enterprise data platform (MSKÜ) — Project lead for a university-wide platform: data dictionary → SQL data model → automated reporting dashboard for every academic and administrative unit.
- MSc thesis — Brain tumour detection with image processing and deep learning.
- MSstatsQC — Contributing to the Bioconductor package for QC of mass-spectrometry proteomics: integrating a new statistical method and the recently published MSstatsQC-ML supervised learning approach.
| Project | What it is |
|---|---|
| student-clustering-dashboard | LEANCLUST — R/Shiny dashboard clustering students weekly via DTW + Ward.D2 to surface at-risk learners. Paper under review at Journal of Learning Analytics (first author). |
| netflix-prize-data-analysis | Recommender system over 100M+ ratings. 0.81 RMSE in 300 seconds — better than the 0.8567 that won the original Netflix Prize. |
| wineapp | End-to-end MLOps: predicting Bordeaux wine scores from text reviews, deployed as a live Shiny app. |
| R | R for Everyone — an open collection of worked examples covering R packages, statistical methods, and ML workflows. |
| bordeaux_wine | Analysis of chemical and sensory drivers of wine quality. |
📕 Gürünlü Alma, Ö., Şafak, S., & Özyurt, U. (2026). Machine Learning Based Missing Value, Outlier and Dimensionality Reduction Methods — With R Applications. Nobel Academic Publishing. ISBN 978-625-386-811-6
📄 Şafak, S., Adnan, M., Doğu, E., & Akhanlı, S. E. LEANCLUST: Improving Instructional Decision-Making via Longitudinal Analytics and Dynamic Student Profiling. Journal of Learning Analytics — under review.
Languages R · Python · SQL · PL/SQL · Java · C
BI & Viz Power BI · Tableau · R Shiny · Oracle Data Integrator
ML Classification · Clustering · Dimensionality reduction · Imputation · Recommender systems · Image processing
Stats SPSS · Minitab · Jamovi · EViews · Weka
LinkedIn · setosfk@outlook.com · Muğla, Türkiye