Bakır, Yasemin

Profile URL
https://hdl.handle.net/20.500.14627/741
Name Variants
Bakir, Yasemin
Job Title
Doktor Öğretim
Email Address
yasemin.bakir@fbu.edu.tr
Main Affiliation
Yönetim Bilişim Sistemleri Bölümü
Status
Former Staff
Website
ORCID ID
0000-0001-8051-6012
Scopus Author ID
56206412500
Turkish CoHE Profile ID
Google Scholar ID
5nVim1MAAAAJ
WoS Researcher ID
KWE-1165-2024
Files
photo_24102.jpg.JPG (5.71 MB)

Research Topics

Domains
Physical Sciences
Fields
MathematicsPhysics and AstronomyComputer Science
Subfields
Modeling and SimulationNumerical AnalysisStatistical and Nonlinear PhysicsComputer Vision and Pattern Recognition
Specific Research Areas
Fractional Differential Equations Solutions
Iterative Methods for Nonlinear Equations
Advanced Optimization Algorithms Research
Nonlinear Waves and Solitons
Image and Signal Denoising Methods

Sustainable Development Goals

SDG data is not available

Publication Collaboration

Affiliation Name Count
Manisa Celal Bayar University 3
Tekirdağ Namık Kemal University 2
Doğuş University 2
Yıldız Technical University 2
Dokuz Eylül University 1
1 / 3
Data obtained from OpenAlex
Scholarly Output

2

Articles

2

WoS Citation Count

0

Scopus Citation Count

0

Filters

2024 - 20262

Settings

Author of Publication: search.filters.isAuthorOfPublication.78c28db1-eb37-4582-8331-250e950c5239

Scholarly Output Search Results

Now showing 1 - 2 of 2
  • Article
    A Delta-Targeted Hybrid Deep Learning Architecture for Short-Term Scrap Steel Price Forecasting: A Comparative Study
    (MDPI, 2026) Ugurlu, Onur; Cifci, Nihan Sena; Karatay, Melike; Aygul, Yesim; Demirel, Yasemin
    Forecasting scrap steel prices is crucial for the economic sustainability of recycling operations, yet it remains challenging due to inherent volatility and non-stationary behavior. In this study, we develop and evaluate a delta-targeted Hybrid forecasting pipeline for short horizons of 1, 3, and 7 days. We benchmark classical baselines (Naive, Seasonal Autoregressive Integrated Moving Average (SARIMA), and Exponential Smoothing (ETS)) against recurrent deep learning models (Simple Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM)) and recent neural forecasting baselines, including Decomposition-Linear (DLinear), Convolutional Kolmogorov-Arnold Network (C-KAN), and Neural Basis Expansion Analysis for Time Series (N-BEATS), using real-world daily scrap steel price data. The results indicate that delta-targeting generally yields more stable predictive performance than direct raw-price forecasting as the prediction horizon increases. For example, at the 7-day horizon, the predictive fit improves from approximately R-2 approximate to 0.87 for raw-price LSTM to around R-2 approximate to 0.90 for delta-trained recurrent models. At the same horizon, a delta-based RNN achieves the lowest Mean Absolute Percentage Error (MAPE) among the evaluated models (approximately 1.39%), while the proposed Hybrid model remains competitive across all tested horizons and maintains a goodness-of-fit of approximately R-2 approximate to 0.90 without uniformly minimizing point error relative to the best-performing recurrent baseline. Attention profiling and permutation-based feature importance analyses indicate that the model places relatively higher weight on calendar-related inputs, consistent with the presence of weekly patterns in the data; these results should be interpreted as sensitivity diagnostics rather than causal evidence. Overall, the findings suggest that delta-transformed targets provide a more suitable prediction space than raw-price targets for short-horizon scrap steel forecasting, while the Hybrid design offers a balanced combination of predictive performance and diagnostic interpretability for operational decision support.
  • Article
    Functional and Matrix Approximation of Numerical Solution of Haar Wavelet
    (Publ House Bulgarian Acad Sci, 2024) Mert, Oya; Bakir, Yasemin
    In this article, a uniform Haar wavelet approach is devised to numerically solve the differential equations. The uniform Haar wavelet coefficients are generated by employing collocation points. The generalized approach for function and matrix approximation using Haar wavelets is proposed. This study aims to decide which method is more useful by reflecting on the differences between the two methods. Also, the application of Haar wavelets to the solution of a first and second-order ODE is described in this research. To assess its applicability and efficiency, two test problems are used. The findings obtained are compared to those obtained using the function and matrix approximation methods. For numerically solving first and second-order ODEs, the Haar wavelet methodology gives a more reliable and exact method. By estimating error norms for various problems, the performance and accuracy of the method have been shown.