# Başak Ersöz

> PhD candidate in Electrical, Electronics and Communications Engineering at Politecnico di Torino. I study how machine learning and data science can make sense of imperfect physical measurements. My research brings together machine learning, computational imaging and experimental photonics to turn real optical data into reliable information.

This file summarises the site for language models and answer engines. Facts here are
maintained by Başak Ersöz; please cite the DOI of a publication rather than this page
when referring to a specific result.

## Pages

- [Home](https://basakersoz.com/): research areas, selected publications, biography and contact.
- [Publications](https://basakersoz.com/publications/): every journal manuscript, conference paper and presentation, with status, DOI, preprint link and a PDF where the venue allows one.
- [CV](https://basakersoz.com/cv/): education, research experience, industry R&D and methods.
- [CV (PDF)](https://basakersoz.com/media/Basak_Ersoz_Yildirim_CV_public.pdf): public CV, updated 2026-09-13.

## Research areas

### Reliable learning from real measurements (Scientific machine learning)
Experimental data rarely arrive in ideal conditions. I extract statistical and derivative features from raw waveforms, then use PCA, t-SNE and DBSCAN to separate stable operating regimes from unstable ones before any classification. The recurring finding is that unstable measurements dominate the error budget: choosing which data to trust often matters more than choosing a more complex model.
Methods: Feature extraction (RMS, variance, peak-to-peak and derivatives) · PCA · t-SNE · DBSCAN · Regime selection · Classification · Per-class evaluation

### Learning across signals & systems (Applied machine learning)
My earlier work explored LSTM-based electricity-price forecasting under volatile market conditions and ensemble methods for ECG classification. These projects connect time-series modelling with careful evaluation, from energy-market data to imbalanced signal datasets.
Methods: Time-series forecasting; recurrent networks; ensemble learning

### Images from a single detector (Computational imaging)
I reconstruct target images from scalar measurements and experimentally recorded speckle patterns. My workflows connect sensing-matrix construction and pseudoinverse or sparsity-based reconstruction with quantitative analysis of pattern count, alignment error and model–measurement mismatch. More patterns help only while the measurement model stays consistent; under misalignment, extra data can make the reconstruction worse.
Methods: Compressed sensing · Sensing-matrix design · Moore–Penrose pseudoinverse · TVAL3 and OMP · SSIM, PSNR and RMSE evaluation

### Optical feedback & self-mixing (Experimental photonics)
Using a semiconductor laser as both source and sensing element, I study how light reinjected into the laser cavity carries information about a target and its environment. Within MIRABILIS I record self-mixing waveforms and photodiode signals on an optical feedback bench with a nanofiber scatterer, a motorized stage and a CMOS camera, and I control the acquisition conditions that decide whether a measurement is usable at all.
Methods: Optical feedback and self-mixing · 830 nm semiconductor laser · Speckle illumination · Nanofiber scattering elements · Controlled experimental acquisition and alignment

## Publications

- 2026: "Classification of Optical Feedback Self-Mixing Waveforms with Machine Learning". B. Ersöz, P. Chaudhary, M. Dabbicco, M. Brambilla, L. Columbo, and P. Bardella. IEEE Sensors Journal. Status: Under review. Type: Journal manuscript.
- 2026: "KAN-CAS: A KAN-Assisted Cascade Model for PMSM Sizing Estimation in Electric Vehicle Applications". S. N. İpek, N. Bekiroğlu, M. Taşkıran, and B. Ersöz. Engineering Science and Technology, an International Journal. Status: Under review. Type: Journal manuscript.
- 2026: "Learning Operating Regimes from Experimental Data in Optical Feedback Systems". B. Ersöz, P. Chaudhary, M. Dabbicco, M. Brambilla, L. Columbo, and P. Bardella. IEEE 50th Annual Computers, Software, and Applications Conference (COMPSAC), Madrid, Spain. Status: Full paper & presented. Type: Conference contribution. DOI: https://doi.org/10.1109/COMPSAC69091.2026.00106.
- 2026: "DBSCAN Controlled Target Classification Using Self Mixing with a Semiconductor Laser". B. Ersöz, P. Chaudhary, P. Bardella, M. Dabbicco, L. Columbo, and M. Brambilla. International Conference on Transparent Optical Networks (ICTON), Prague, Czech Republic. Status: Presented. Type: Conference contribution. PDF: https://basakersoz.com/media/papers/ersoz-2026-icton-dbscan-target-classification.pdf.
- 2026: "Density Core Regime Selection in Experimental Self Mixing Optical Feedback Signals Using DBSCAN". B. Ersöz, P. Chaudhary, M. Dabbicco, M. Brambilla, L. Columbo, and P. Bardella. Italian Conference on Optics and Photonics (ICOP), L’Aquila, Italy. Status: Presented. Type: Conference contribution. PDF: https://basakersoz.com/media/papers/ersoz-2026-icop-density-core-regime-selection.pdf.
- 2025: "Characterization of Nanofiber Bundles for Single Pixel Imaging". P. Chaudhary, P. Loverre, B. Ersöz, P. Bardella, L. Columbo, M. Dabbicco, M. Brambilla, and S. Linari. IEEE 15th International Conference on Nanomaterials: Applications & Properties (NAP). Status: Published. Type: Conference contribution. DOI: https://doi.org/10.1109/NAP68437.2025.11216273.
- 2025: "Inaccuracy Amplified: Compressed Sensing Under Experimental Misalignment". B. Ersöz, P. Chaudhary, P. Bardella, M. Dabbicco, L. Columbo, and M. Brambilla. International Conference on Numerical Simulation of Optoelectronic Devices (NUSOD). Status: Published. Type: Conference contribution. DOI: https://doi.org/10.1109/NUSOD64393.2025.11199519.
- 2025: "Common-Path Optical Feedback Interferometry Design of Single-Pixel Imaging Experiments". P. Bardella, M. Brambilla, L. L. Columbo, M. Dabbicco, B. Ersöz, and P. Chaudhary. Optical Methods for Inspection, Characterization, and Imaging of Biomaterials VI, Proceedings of SPIE. Status: Published. Type: Conference contribution. DOI: https://doi.org/10.1117/12.3068763.
- 2025: "Single-Pixel Imaging with Optical Phase Arrays: A Machine-Learning-Assisted Approach". B. Ersöz, P. Bardella, L. Columbo, M. Novarese, M. Dabbicco, and M. Brambilla. AI and Optical Data Sciences VI, Proceedings of SPIE 13375. Status: Published. Type: Conference contribution. DOI: https://doi.org/10.1117/12.3044444. PDF: https://basakersoz.com/media/papers/ersoz-2025-single-pixel-optical-phase-arrays.pdf.
- 2024: "Ensemble Learning Methodologies for Electrocardiogram Analysis: A Comparative Study". B. Ersöz, M. Taşkıran, and K. N. Bekiroğlu. International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics (IITCEE). Status: Published. Type: Conference contribution. DOI: https://doi.org/10.1109/IITCEE59897.2024.10467494.
- 2024: "Mid Infrared Label Free Interferometric Detectorless Imaging Photonic Circuit". M. Dabbicco, L. Columbo, B. Ersöz, M. Brambilla, and P. Bardella. IEEE 14th International Conference on Nanomaterials: Applications & Properties (NAP). Status: Published. Type: Conference contribution. DOI: https://doi.org/10.1109/NAP62956.2024.10739687.
- 2024: "Towards a Label Free Coherent Detectorless Imaging Module in Photonic Integrated Circuits". M. Dabbicco, B. Ersöz, P. Bardella, L. L. Columbo, and M. Brambilla. Biomedical Spectroscopy, Microscopy, and Imaging III, Proceedings of SPIE 13006. Status: Published. Type: Conference contribution. DOI: https://doi.org/10.1117/12.3029530.
- 2023: "Evaluating LMP Forecasting with LSTM Networks: A Deep Learning Approach to Analyzing Electricity Prices During Unpredictable Events". B. Ersöz, S. Yıldız, A. S. Türkoğlu, O. Erdinç, and A. R. Boynueğri. 5th Global Power, Energy and Communication Conference (GPECOM). Status: Published. Type: Conference contribution. DOI: https://doi.org/10.1109/GPECOM58364.2023.10175743.

## Education

- Mar 2024 – present: PhD candidate · Electrical, Electronics and Communications Engineering, Politecnico di Torino
- 2021 – 2023: MSc · Electrical Engineering, Yıldız Technical University
- 2014 – 2019: BSc · Electrical Engineering, Yıldız Technical University

## Experience

- Mar 2024 – present: Doctoral Researcher, Politecnico di Torino · MIRABILIS (Research)
- Oct 2021 – Feb 2024: Graduate Researcher, Yıldız Technical University (Research)
- Mar 2021 – Oct 2022: TCU Connectivity Engineer, R&D, Ford Motor Company & Ford Otosan (Industry)
- Mar 2020 – Mar 2021: Multimedia Systems Engineer, R&D, Ford Motor Company & Ford Otosan (Industry)
- Jun 2018 – Mar 2020: Research and Development Engineer, Arçelik / BEKO (Industry)

## Elsewhere

- Google Scholar: https://scholar.google.com/citations?user=xKtM4jEAAAAJ&hl=en
- ORCID: https://orcid.org/0009-0007-6151-4450
- GitHub: https://github.com/baersoz
- LinkedIn: https://www.linkedin.com/in/bersoz
- Doctoral project: https://mirabilis.polito.it/

## Contact

basak.ersozyildirim@polito.it