Bilgisayar Mühendisliği Bölümü Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.11779/1940
Browse
50 results
Search Results
Conference Object Churn Prediction for Subscription-Based Applications Using Machine Learning(Institute of Electrical and Electronics Engineers Inc., 2025-09-17) Gozukara H.; Patel J.; Kara E.; Yildiz A.; Mese Y.K.; Obali E.; Cakar T.; Gozukara, Hamza; Mese, Yasemin Kurtcu; Patel, Jay; Kara, Erkan; Yildiz, Aysenur; Cakar, Tuna; Obali, EmirIn this study, a predictive model was developed using machine learning techniques to forecast customer churn in subscription-based video streaming services. The data such as user login records, content viewing information, subscription details, and content-related features were used to interpret usage patterns and customer churn was defined based on subscription renewal status and renewal timing. Several usage-based features are extracted for users and several algorithms were used for modeling, such as Random Forest, CatBoost, XGBoost, Logistic Regression, K-Nearest Neighbors, and Gradient Boosting. Occurring class imbalance on the target variable is handled via BorderLineSMOTE. The model's performance was evaluated using training-test accuracy plots, classification reports, and hyperparameter tuning. Even though most of the models performed similarly, the CatBoost model emerged as the top performer, achieving a macro F1-score of 0.60. However, while effective in identifying churners, the models struggled to precisely classify non-churning customers, a common challenge in imbalanced datasets even after applying oversampling techniques. The analysis of feature importance yielded a crucial insight, early and consistent user engagement is the strongest predictor of customer retention. These findings provide valuable, actionable insights for streaming platforms, emphasizing that retention strategies should focus on maximizing engagement immediately after a user subscribes. © 2025 IEEE.Conference Object Attention-Enhanced Dual-Head LSTM With Rich Feature Engineering for Risk-Adjusted Stock Return Forecasting(Institute of Electrical and Electronics Engineers Inc., 2025-09-17) Patel J.; Gunes P.; Ertugrul S.; Sayar A.; Benli H.; Makaroglu D.; Cakar T.; Benli, Harun; Gunes, Peri; Patel, Jay; Makaroglu, Didem; Sayar, Alperen; Cakar, Tuna; Ertugrul, SeyitStock return forecasting is a challenging task due to the complex, nonlinear, and volatile nature of financial markets. In this paper, we propose a comprehensive deep learning framework that integrates: a two-layer Long Short-Term Memory (LSTM) network augmented with a learnable attention mechanism, a dual-head output for simultaneous regression of next-day returns and classification of price direction, with an extensive suite of technical and macro-financial features. Our feature set comprises lagged log-returns, trend indicators (simple and exponential moving averages), momentum oscillators (RSI, MACD), volatility measures (rolling variance and GARCH conditional volatility), price bands (Bollinger Bands, Donchian channels), volume metrics (On-Balance Volume, Volume Rate of Change), Hidden Markov Model regime states, market index returns, and calendar effects. We train and validate the model using a rolling-window cross-validation scheme with early stopping and hyperparameter tuning to ensure temporal robustness. Empirical results on a large multi-stock dataset demonstrate that our attention-enhanced, dual-task LSTM outperforms single-task LSTMs and traditional machine learning benchmarks, achieving lower forecasting error and more stable generalization. © 2025 IEEE.Conference Object Rag Based Interactive Chatbot for Video Streaming Services(Institute of Electrical and Electronics Engineers Inc., 2025-09-17) Gözükara H.; Patel J.; Kara E.; Yildiz A.; Köseoǧlu O.; Makaroǧlu D.; Çakar T.; Gözükara, Hamza; Patel, Jay; Makaroǧlu, Didem; Kara, Erkan; Yildiz, Ayşenur; Çakar, Tuna; Köseoǧlu, OzanThe proliferation of content within video streaming services presents a significant challenge for users seeking personalized recommendations and specific information. This research addresses this challenge by developing a Retrieval-Augmented Generation (RAG) chatbotn designed to enhance user experience through conversational AI. The primary contribution of this work is a novel Retrieval-Augmented Generation (RAG) architecture featuring a dual-retrieval system that combines semantic search for descriptive requests and structured queries for fact based inquiries. This approach grounds the Large Language Model (LLM) in a factual knowledge base, mitigating the risk of hallucinations. The system is engineered to handle empty data retrieval scenarios by dynamically relaxing search filters, ensuring a robust user experience. The effectiveness of this RAG approach was validated through a comprehensive set of automated evaluations. The system demonstrates high precision in ranked list retrieval with questions like "Recommend me the top 5 action movies with highest IMDb scores", achieving an average NDCG@k of 0.837. While the chatbot shows strong semantic understanding by achieving 91% accuracy with contextual clues such as "Which Batman movies are directed by Christopher Nolan?", its performance with more ambiguous, plot-only queries (59.5% accuracy) indicates clear opportunities for future refinement. These results confirm that the dual-tool architecture successfully combines the flexibility of semantic search with the precision of structured queries, paving the way for more intuitive and efficient content discovery on streaming platforms. © 2025 IEEE.Conference Object Combining Similar Trajectories and XGBoost via Residual Learning for Traffic Flow Forecasting(Institute of Electrical and Electronics Engineers Inc., 2025-11-14) Işlak U.; Yilmaz E.; Arslan I.; Çakar T.; Çakar, Tuna; Işlak, Ümit; Yilmaz, Elif; Arslan, IlkerIn this study, we propose novel hybrid forecasting models that integrate the method of similar trajectories with machine learning techniques, particularly the XGBoost algorithm, for traffic flow prediction. Traditional statistical models, such as ARIMA, often struggle to accurately capture the complex, non-linear patterns characteristic of traffic flow data. To address these limitations, we develop an additive hybrid forecasting framework that combines the strengths of linear models (similar trajectories method) and non-linear models (XGBoost). Our proposed methods are evaluated on two different stations from the California PEMS dataset. Experimental results demonstrate that the proposed hybrid models consistently outperform individual benchmark models, including ARIMA, standalone similar trajectories, and XGBoost. The superiority of the hybrid approach, particularly the XGBST model, is further validated through the Diebold-Mariano statistical test, confirming significant predictive improvements at various significance levels. Additionally, using weighted Euclidean distance within the similar trajectories method further enhanced forecasting accuracy. The interpretability and flexibility of our hybrid framework make it especially suitable for practical implementation in traffic management systems. These findings underline the effectiveness of hybrid modeling strategies in traffic flow forecasting and suggest future research directions, such as comprehensive hyperparameter optimization and broader validation across diverse datasets. © 2025 IEEE.Conference Object A Multimodal AI and ML Framework for Fashion Image Segmentation, Recommendation, and Similarity Recognition(Institute of Electrical and Electronics Engineers Inc., 2025-09-17) Soyhan M.E.; Ay T.B.; Memis E.C.; Fatih Capal M.; Cakar T.; Gunay S.; Coskun H.; Gunay, Savas; Memis, Emir Cetin; Fatih Capal, Mehmet; Soyhan, Mustafa Eren; Coskun, Hasan; Cakar, Tuna; Ay, Tarik BugraThis study presents a scalable multimodal Artificial Intelligence (AI) and Machine Learning (ML) framework designed to enhance decision making in the fashion industry. The proposed system integrates garment segmentation, multimodal feature extraction, and similarity recommendation into a unified pipeline. Using Segformer for segmentation, along with the convolutional neural network (CNN)-based feature extraction models ResNet152V2 and Xception, and the transformer-based vision-language model LLaVA, the framework generates visual and semantic embeddings of garments. These representations are processed through similarity detection using OpenAI embedding models and stored in the Pinecone vector database for efficient retrieval. Real-time similarity scoring is enabled through FastAPI endpoints, offering interactive search capabilities. Preliminary results demonstrate the system's strong ability to identify conceptually and visually similar items across a large catalog, providing actionable insights for designers. This framework lays the groundwork for intelligent, interpretable, and production-ready AI systems in the fashion industry. © 2025 IEEE.Conference Object Graph Theory-Based Fraud Detection in Banking Check Transactions(Institute of Electrical and Electronics Engineers Inc., 2025-09-17) Behsi Z.; Memis E.C.; Ertugrul S.; Sayar A.; Gunes P.; Seydioglu S.; Cakar T.; Gunes, Peri; Memis, Emir Cetin; Sayar, Alperen; Cakar, Tuna; Ertugrul, Seyit; Seydioglu, Sarper; Behsi, ZeynepTraditional banking fraud detection systems rely on rule-based approaches that analyze individual transactions in isolation, failing to capture complex relationship patterns indicative of coordinated fraud schemes such as check-kiting and artificial credit score manipulation. We p resent our study, a novel similarity-based graph theory approach that constructs weighted networks between check issuers using Jaccard Similarity Index and employs advanced graph analysis to identify suspicious entity clusters without requiring complete transaction relationship data. Our approach combines Jaccard Similarity Index for behavioral pattern analysis (addressing payee information unavailability) with comprehensive graph analysis including centrality measures, community detection, and anomaly identification. Through comprehensive evaluation on real banking data containing 458,399 transactions from 121,647 unique issuers - the largest confirmed dataset in fraud detection literature - we demonstrate the effectiveness of our methodology. Following parameter optimization using grid search methodology (similarity threshold: 0.55, risk percentile: 0.75), our study achieves competitive detection rates in optimal configurations with an average F1-score of 0.447 (±0.164) and peak performance reaching an F1-score of 0.557, while providing superior network topology analysis with 0.923 clustering coefficient. The system operates under significant data privacy constraints, lacking personal identification information (names, account numbers, IDs) and complete payee data. Despite these limitations, our study outperforms traditional approaches by leveraging similarity-based indirect relationships, and we project that performance could reach 85-95% levels with complete data access. © 2025 IEEE.Conference Object Predicting Customer Churn in Retail Using Machine Learning on Transaction Data(Institute of Electrical and Electronics Engineers Inc., 2025-09-17) Bozan M.T.; Gozukara H.; Patel J.; Kizilay A.; Sahin Z.; Tosun B.; Cakar T.; Gozukara, Hamza; Kizilay, Ayse; Patel, Jay; Bozan, Mehmet Talha; Cakar, Tuna; Tosun, Busra; Sahin, ZeynepCustomer churn prediction is critical for businesses to retain customers and reduce revenue loss. This paper presents a retail customer churn prediction study. We preprocess transactional data from a retail dataset comprising approximately 19.7 million transactions involving over 1 million customers. Temporal behavioral features, such as purchase frequency, monetary value, product variety, and promotional engagement metrics, are engineered using a four-month observation window. A Random Forest classifier is trained, utilizing balanced class weighting to address churn class imbalance. The churn label is defined as customers not purchasing in the subsequent six-month period. Our Random Forest model achieves approximately 84% accuracy, 86% precision, 85% recall, and an F1- score of 85%. Additionally, an XGBoost model achieves similar accuracy (≈ 84%) but higher recall (93%) and F1-score (89%), indicating improved churn prediction. The confusion matrix illustrates clear model performance. This study demonstrates that carefully engineered RFM-based features and ensemble learning approaches significantly enhance churn prediction in retail contexts. © 2025 IEEE.Conference Object Developing Autonomous Steering Algorithm To Improve Cornering Slip Performance of a Four-Wheel Car Using Neural Network Tools(Institute of Electrical and Electronics Engineers Inc., 2025-09-17) Alatciyan D.R.; Emeryan B.J.; Barbaros B.; Cakar T.; Kilic N.; Emeryan, Burak Jirayr; Kilic, Namik; Alatciyan, Diran Robin; Barbaros, Bugra; Cakar, TunaThis study investigates a neural network-based predictive steering control using simulation data generated from ADAMS Car. A Long Short-Term Memory (LSTM) architecture is employed to estimate steering angle and longitudinal velocity from sequential input features, with the goal of analyzing the model's behavior in cornering scenarios. The experimental setup includes multiple simulation runs under varying configurations, particularly exploring the effect of different sliding window sizes on prediction performance. Results show that the proposed model can effectively capture temporal patterns in the input data and produce consistent estimations across test conditions. While the study is limited to a simulation environment, it provides initial insights into how AI-based models may support steering control tasks and lays the groundwork for future extensions involving additional vehicle dynamics inputs. © 2025 IEEE.Conference Object Citation - WoS: 3Citation - Scopus: 3Detecting Autism From Head Movements Using Kinesics(Assoc Computing Machinery, 2024-11-04) Gokmen, Muhittin; Sariyanidi, Evangelos; Yankowitz, Lisa; Zampella, Casey J.; Schultz, Robert T.; Tunc, BirkanHead movements play a crucial role in social interactions. The quantification of communicative movements such as nodding, shaking, orienting, and backchanneling is significant in behavioral and mental health research. However, automated localization of such head movements within videos remains challenging in computer vision due to their arbitrary start and end times, durations, and frequencies. In this work, we introduce a novel and efficient coding system for head movements, grounded in Birdwhistell's kinesics theory, to automatically identify basic head motion units such as nodding and shaking. Our approach first defines the smallest unit of head movement, termed kine, based on the anatomical constraints of the neck and head. We then quantify the location, magnitude, and duration of kines within each angular component of head movement. Through defining possible combinations of identified kines, we define a higher-level construct, kineme, which corresponds to basic head motion units such as nodding and shaking. We validate the proposed framework by predicting autism spectrum disorder (ASD) diagnosis from video recordings of interacting partners. We show that the multi-scale property of the proposed framework provides a significant advantage, as collapsing behavior across temporal scales reduces performance consistently. Finally, we incorporate another fundamental behavioral modality, namely speech, and show that distinguishing between speaking- and listening-time head movements significantly improves ASD classification performance.Conference Object Predicting Credit Repayment Capacity With Machine Learning Models(Ieee, 2024-05-15) Filiz, Gozde; Bodur, Tolga; Yaslidag, Nihal; Sayar, Alperen; Çakar, TunaThis study examines the transformation in the financial services sector, particularly in banking, driven by the rapid development of technology and the widespread use of big data, and its impact on credit prediction processes. The developed credit prediction model aims to more accurately predict customers' credit repayment capacities. In pursuit of this goal, demographic and financial data along with credit histories of customers have been utilized to employ data preprocessing techniques and test various classification algorithms. Findings indicate that models developed with XGBoost and CATBoost algorithms exhibit the highest performance, while the effective use of feature engineering techniques is revealed to enhance the model's accuracy and reliability. The research highlights the potential for financial institutions to gain a competitive advantage in risk management and customer relationship management by leveraging machine learning models.