Page Not Found
Page not found. Your pixels are in another canvas.
Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Page not in menu
This is a page not in th emain menu
About me
About me
Posts
Hike
Published:
Living in California has given me the chance to explore some of the most beautiful trails and landscapes. From coastal cliffs to the top of the waterfalls, hiking has become one of my favorite ways to recharge and enjoy nature. Yosemite National Park, CA Sequioa National Park, CA Garnet Peak, CA
Guitar and Me
Published:
I have been playing guitar since 2012, and it has been one of my favorite creative outlets. Over the years, I’ve had the chance to perform as part of a guitar duo within an orchestra (harder than it sounds!). Each performance was a memorable experience, blending practice, collaboration, and the joy of live music.
portfolio
Deep Learning for Self Driving Cars
Deep Learning for Self Driving Cars — Department of Computer Science and Engineering, UC San Diego.
ChatGPT Analysis with STS
ChatGPT Analysis with STS — Department of Electrical and Electronics Engineering, Bilkent University.
Geolocation Source Localization
Geolocation Source Localization — ASELSAN, Inc.
A Basic Level Category Analysis with Commonsense Question Answering
A Basic Level Category Analysis with Commonsense Question Answering — Department of Electrical and Electronics Engineering, Bilkent University.
Digital FPGA Piano for Beginners
Digital FPGA Piano for Beginners — Bilkent University
Recommendify : Song Recommendation System for Spotify Playlists
Recommendify : Song Recommendation System for Spotify Playlists — Department of Computer Science, Bilkent University.
Machine Learning Based Spatio-Temporal Prediction System for Traffic Accidents
Machine Learning Based Spatio-Temporal Prediction System for Traffic Accidents — Department of Electrical and Electronics Engineering, Bilkent University.
Wind Energy Production Prediction
Wind Energy Production Prediction — Department of Electrical and Electronics Engineering, Bilkent University.
A forecasting system for total electric consumption plays a prominent role in the electric generation market. Moreover, creating a precise forecast system is a hot topic for many high-tech companies. To this end, our goal is to forecast hourly total electrical energy consumption in Spain with regression models, which are Linear Regression, Decision Tree, and AdaBoost. The machine learning models are coded without library support.
publications
Hierarchical Ensemble-based Feature Selection for Time Series Forecasting
Submitted to Machine Learning, 2024
This paper introduces a novel ensemble approach for feature selection based on hierarchical stacking…
Recommended citation: A. Tumay, M. E. Aydin, A. T. Koc, S. S. Kozat. " Hierarchical Ensemble-based Feature Selection for Time Series Forecasting." Machine Learning. Submitted, 2023. (doi:10.48550/arXiv.2310.17544)
Probabilistic Digital Twin for Data-driven Smart Weaning of Medical Circulatory Devices
Submitted to NeurIPS Learning from Time Series for Health, 2025
Recommended citation: S. H. Sun, A. Tumay, S. Fereidooni, A. Dumas, E. Jortberg, R. Yu. “Probabilistic Digital Twin for Data-driven Smart Weaning of Medical Circulatory Devices.” NeurIPS Learning from Time Series for Health 2025 (Under review).
Guardian-regularized Safe Offline Reinforcement Learning for Smart Weaning of Mechanical Circulatory Devices
Submitted to Machine Learning for Health, 2025
Recommended citation: A. Tumay, S. H. Sun, S. Fereidooni, A. Dumas, E. Jortberg, R. Yu. “Guardian-regularized Safe Offline Reinforcement Learning for Smart Weaning of Mechanical Circulatory Devices.” Machine Learning for Health 2025 (Under review).
talks
5th Place in Data Analysis Challange
Published:
Seasonal demand planning is essential for retailers as it provides a strategic approach to meeting demand throughout the season, considering factors like promotions, changing customer preferences, and new product introductions. Despite its complexity, effective seasonal demand forecasting helps prevent both stock-outs, which result in financial losses, and excess stock, leading to storage costs and product expirations. Bogazici University Operational Research Club challenges competitors to create a highly accurate product forecasting model for seasonal needs.
3rd Place in IPSOS Datathon
Published:
The market share of a food company is forecasted for 2020-2022 based on the previous 6 years of data with ARIMA, and linear regression. The importance of other companies’ market share for predicting this company’s market share is investigated. The marketing strategies for increasing the market share of the sub-company of this food company are investigated in terms of customer parameters such as closeness, top of mind, etc.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.