01. Araştırma Çıktıları | WoS | Scopus | TR-Dizin | PubMed

Permanent URI for this communityhttps://hdl.handle.net/20.500.11779/255

Browse

Now showing 1 - 7 of 7

Dil Modelleri ile Akademik Özet Üretimi
(Institute of Electrical and Electronics Engineers Inc., 2025) Bektas, Busra; Gultekin, Ali Ozgun; Ozdemiroglu, Emre; Yilmaz, Zeynep; Dikici, Buse; Demir, Seniz
In recent years, large language models have demonstrated extraordinary capabilities in natural language processing tasks. The integration of these models to text summarization has highlighted the need for evaluating varying model performances under a standardized benchmarking framework. In this study, the performance of different large language models in generating abstracts of scientific papers which has a common structure and unique language is compared through an extensive experimental analysis. The abstracts automatically generated by these models using prompt engineering were evaluated via various evaluation metrics based on content overlap and semantic similarity. The results that we obtained demonstrated the effectiveness of large language models in abstract generation. © 2025 Elsevier B.V., All rights reserved.
Ensemble-Based Stock Prediction for Retail - XGBoost and LightGBM with Rolling Window Training
(Institute of Electrical and Electronics Engineers Inc., 2025) Patel, Jay Nimish; Kizilay, Ayse; Şahin, Zeynep; Sercan, Busra; Toprak, Samet; Çakar, Tuna
Stock prediction in retail settings is a critical challenge that impacts numerous businesses globally, that require precise and timely forecasts to optimize inventory management and enhance customer satisfaction. State-of-the-art approaches for accurate stock prediction leverage machine learning (ML) models, which require large amounts of historical sales data for effective training. Such detailed datasets are often hard to obtain, limiting the performance and scalability of these approaches. In this paper, we propose various strategies to tackle this limitation. Initially, we adopt a transfer-learning approach, utilizing pre-trained models like XGBoost and LightGBM, which are fine-tuned for stock prediction in retail environments. To further boost model performance, we incorporate an ensemble method that combines predictions from both models to improve accuracy and manage outliers. Experiments conducted on an extremely large dataset, comprising millions of retail transactions, highlight the presence of significant outliers. Our models, augmented with ensemble strategies, significantly outperform traditional models in handling these complexities and improving prediction accuracy. © 2025 Elsevier B.V., All rights reserved.
Frekans Seçici Yüzeyler ile Mikroşerit Yama Antenin Radar Kesit Alanının Azaltılması
(Institute of Electrical and Electronics Engineers Inc., 2025) Metin, Ahmet Resul; Aydin, Irem; Bilgin, Egemen
In this study, the radar cross section (RCS) of a microstrip patch antenna was reduced via frequency selective surfaces (FSS). Since microstrip antennas are frequently used in military platforms, antenna design with low radar cross section will contribute to the general efforts towards reducing RCA. First, a microstrip antenna with a center frequency of 5 GHz was designed and a FSS that acts as a reflector at this frequency was developed. Here, the expectation from the FSS is to act as a reflector at the center frequency and to have a transmissive behavior at other frequencies, contributing to the reduction of the radar cross section. By integrating the designed FSS structure into the ground plane of the antenna, RCA reduction was achieved in different bands without shifting the center frequency. According to the simulations, the designed antenna has a 10 dB lower radar cross section in the X band. © 2025 Elsevier B.V., All rights reserved.
İnternet Trafik Hızının Tahmininde Derin Öğrenme ve Ağaç Tabanlı Modellerin Karşılaştırılması
(Institute of Electrical and Electronics Engineers Inc., 2025) Filiz, Gozde; Altıntaş, Suat; Yıldız, Ayşenur; Kara, Erkan; Drias, Yassine; Çakar, Tuna
This study addresses the prediction of internet traffic speed using time-dependent data from an internet service provider through different modeling approaches. On an anonymized dataset, the performance of the moving average method, various deep learning models (N-BEATS, N-HITS, TimesNet, TSMixer, LSTM), and the XGBoost regression model enhanced with feature engineering was compared. Time series cross-validation and random hyperparameter search were used for model training. According to the results, the XGBoost model achieved the highest accuracy with 98.7% explained variance (R2), while among the deep learning models, N-BEATS and N-HITS achieved the best performance with R2 values around 90%. The findings indicate that tree-based methods supported by carefully selected features can offer higher accuracy and computational efficiency compared to complex deep learning models in internet traffic forecasting. © 2025 Elsevier B.V., All rights reserved.
Makine Öğrenimi ve Çok Boyutlu Anket Verileri Kullanılarak Öğrenci Başarısının Tahmini: Eğitim Programı Üzerine Bir Uygulama
(Institute of Electrical and Electronics Engineers Inc., 2025) Behsi, Zeynep; Dereli, Serhan; Çakar, Tuna ; Patel, Jay Nimish; Cicek, Gultekin; Drias, Yassine
This study develops a machine learning model integrating survey data and performance metrics to predict student success in the UpSchool education program. Students' personality traits assessed by DISC analysis, financial management, social, and emotional skills were clustered into "Successful,""Unsuccessful,"and "Moderately Successful"groups using K-means clustering. The SMOTE technique addressed data imbalance issues, and algorithms such as Logistic Regression, Random Forest, LightGBM, and AdaBoost were tested. After hyperparameter optimization, AdaBoost and LightGBM achieved the highest predictive performance. Results demonstrated the effectiveness of machine learning models in forecasting student success in educational programs. Future studies are recommended to enhance model performance through expanded datasets and to validate the model's applicability across diverse educational contexts. © 2025 Elsevier B.V., All rights reserved.
Mikrodalga Tomografi ile Doku İçindeki Yabancı Unsurların Konumlandırılması
(Institute of Electrical and Electronics Engineers Inc., 2025) Aydin, Pelinsu; Guler, Gokcenaz; Togar, Ayca; Bilgin, Egemen
The aim of medical imaging systems is to detect harmful elements such as tumors in biological tissues via non-invasive methods. To this end, a low-cost and unharmful microwave tomography system is proposed in this study. S-parameters, which are related to the scattered field, are measured with wide-band Vivaldi antennas placed around the examined tissue. These parameters are calibrated by analytical scattered field to increase data quality. Then, an inverse scattering problem is formulated using them as data and solved via Reverse Time Migration (RTM) and Multiple Signal Classification (MUSIC) methods. This solution produces an indicator function indicating the location of the foreign element within the tissue. The proposed method has been tested with simulations in the CST environment. The results show that the system can successfully detect single or multiple foreign substances in the tissue. © 2025 Elsevier B.V., All rights reserved.
Yapay Öğrenme Tabanlı Mikrofaktoring Skorlama Modeli ve Kredi Risk Yönetim Sistemi Geliştirilmesi
(Institute of Electrical and Electronics Engineers Inc., 2025) Sayar, Alperen; Ates, Yigit; Ertugrul, Seyit; Turan, Elif Naz; Drias, Yassine; Çakar, Tuna
Credit scoring systems are critical tools used by factoring institutions to assess the credit risks of SME businesses seeking microloans. This study presents a comprehensive predictive modeling framework that achieves 82.67% ROC-AUC with 65.34% Gini score on test data, demonstrating robust discriminative capability despite significant class imbalance. Our ensemble approach outperforms individual boosting models by leveraging their complementary strengths in payment behavior analysis and fraud detection. The raw data was cleaned, transformed, and optimized using the Polars library, with specialized features for detecting fraud patterns and time-based risk indicators. When implementing a score threshold of 950, our model significantly improves the detection of non-performing loans (NPL) compared to traditional rule-based approaches by reducing the net deficit from 6.59% to 2.62%. When applied to previously rejected applications, the model projects a potential 762.57% increase in transaction count and 747.05% growth in transaction volume. © 2025 Elsevier B.V., All rights reserved.

Browse

Browsing 01. Araştırma Çıktıları | WoS | Scopus | TR-Dizin | PubMed by Journal "-- 33rd IEEE Conference on Signal Processing and Communications Applications, SIU 2025 -- Istanbul; Isik University Sile Campus -- 211450"