Yazar "Duman, Mustafa Ozan" seçeneğine göre listele

Listeleniyor 1 - 7 / 7
  • Küçük Resim Yok
    Öğe
    Assessment of an Optimal Multilayer Perceptron Model in Estimation of Stock's Price Rates; A Sample from the Turkish Stock Exchange BIST
    (Institute of Electrical and Electronics Engineers Inc., 2024) Duman, Mustafa Ozan; Tagluk, Mehmet Emin
    Stock price prediction (SPP) is critical for both investors and financial experts to make accurate choices about real estate share transactions. Nowadays, Artificial Neural Networks (ANNs) are commonly used in SPP algorithms. In this study, the performance of a Multi-Layer Perceptron (MLP), a subclass of ANN, in forecasting the closing price of the following day as well as the Direction of Change (DOC) of stock was investigated. The experiments were conducted on four randomly selected stocks from Borsa Istanbul (BIST), using Sigmoid and Rectified Linear Unit (ReLU) activation functions and optimizing with variation of the node numbers in each of the hidden layers of the ANN. For each stock, the data comprised 500 workdays recorded over April 7, 2022, to April 4, 2024 period were used for training and testing the NN model taken under analyses. The predicted values come through ANN were compared with actual data using the Mean Absolute Percentage Error (MAPE) metric. The optimal outcomes were obtained with MLP with ReLU activation function and 8-5-3-1 nodes in the subsequent layers. It was observed that the predictability of stocks changed with macroeconomic and microeconomic factors as well as different time periods, which shows that stock prices also depend on temporal parameters. The overall results showed that stocks' price prediction may not be impossible, but not an easy business. © 2024 IEEE.
  • Küçük Resim Yok
    Öğe
    Bacterial Chemotaxis in Molecular Communication: Experimental and Simulation Analysis of Receiver Placement and Gradient Dynamics
    (Ieee-Inst Electrical Electronics Engineers Inc, 2026) Duman, Mustafa Ozan; Isik, Ibrahim; Isik, Esme
    Bacteria-based nanonetworks (BNs) represent a promising strategy for nanoscale information transfer, utilizing bacterial motility and chemotaxis for targeted message delivery. This study analyzes BN performance through both experimental validation and a custom-developed three-dimensional (3D) simulation program built in MATLAB, focusing on receiver (RX) placement, chemoattractant release rate (Q), and bacterial lifespan. The simulation employs experimentally validated parameters and models bacterial behavior under various spatial configurations. Results demonstrate that RX positioning significantly affects communication efficiency, with asymmetric placement causing uneven chemoattractant gradients and reduced success rates. While higher Q values improve reach time and delivery success, bacterial lifespan becomes a limiting factor at extended distances. Experimental findings using agar-based assays confirm a threshold distance beyond which bacterial motility becomes ineffective. These insights provide practical guidance for optimizing BN systems by balancing signal strength with biological constraints. Future work should explore adaptive bacterial strategies and dynamic environmental conditions to further enhance BN reliability and applicability in areas such as targeted drug delivery and biosensing.
  • Küçük Resim Yok
    Öğe
    Bacterial-Based Molecular Communication: Simulation of a Fixed and Receding Receiver Scenarios in Varied Viscosities and Environmental Conditions
    (Wiley-V C H Verlag Gmbh, 2025) Duman, Mustafa Ozan; Isik, Ibrahim; Isik, Esme; Er, Mehmet Bilal
    This study introduces a novel bacterial-based molecular communication (BBMC) model for nanoscale information exchange, harnessing the chemotactic behavior of Escherichia coli (E. coli). A comprehensive 3D simulation framework is developed to analyze the impact of key parameters diffusion coefficient (D), chemoattractant release rate (Q), receiver (RX) speed (u), and initial transmitter-receiver distance (d) on communication performance. Results indicate that lower D values enhance the formation of chemoattractant gradients, leading to improved signal clarity and efficiency. Conversely, higher RX speeds distort these gradients, increasing signal reach time and reducing success rates. Elevated Q values significantly broaden the sensing range and improve reliability, particularly over larger distances, though their effect is diminished at high RX speeds. Notably, success rates drop sharply as d approaches the theoretical sensing threshold, underscoring the critical need for parameter tuning. Experimental results validate these findings and reveal a threshold beyond which bacterial movement becomes random, limiting effective signal transmission. These insights contribute to optimizing BBMC systems for greater efficiency and reliability. Applications include targeted drug delivery, environmental biosensing, and synthetic biology, where precise bacterial signaling is essential. The study also demonstrates simulation as a scalable, cost-efficient alternative to experimental methods, addressing complexity and feasibility in real-world scenarios.
  • Küçük Resim Yok
    Öğe
    Chemotaxis-Driven Molecular Communication in Nanonetworks: Simulating E. coli Behavior and Performance
    (Institute of Electrical and Electronics Engineers Inc., 2025) Duman, Mustafa Ozan; Isik, Ibrahim; Bilaler, Mehmet; Tagluk, Mehmet Emin; Isik, Esme
    Bacteria-based nanonetworks (BN) show significant potential for revolutionizing nanoscale communication, particularly in fields like medicine and environmental monitoring. This study models the chemotaxis of Escherichia coli (E. coli) in a 2D environment using a customdeveloped MATLAB simulation to understand communication effectiveness. We investigate the impact of chemoattractant release rate (Q), transmitter-receiver distance (d), and bacterial lifespan. Key findings reveal a trade-off between communication range and energy consumption: while higher Q values extend range, they also increase resource usage. A Q value of 10-14 ~mol / s is identified as providing a balanced approach. Furthermore, simulations highlight that bacterial lifespan inherently limits longer communication range, suggesting the potential for nanomachine relays in future BN designs. Future research will expand these models to incorporate 3D environments and multi-bacterium interactions, enhancing their applicability for real-world scenarios. © 2025 IEEE.
  • Küçük Resim Yok
    Öğe
    Performance Assessment of Temporal Frame Size in Stock Price Prediction
    (Institute of Electrical and Electronics Engineers Inc., 2024) Duman, Mustafa Ozan; Tagluk, Mehmet Emin
    Accurate Stock Price Prediction (SPP) is essential for those involved in stock market business. Due to the nonlinear and erratic characteristics of the stock market, estimation of the future prices of stocks is quite challenging. In addition to classical models such as autoregressive integrated moving average (ARIMA), random forest (RF), machine learning-based approaches like Multilayer Perceptron (MLP), Recurrent Neural Network (RNN) possessing Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNN) and their variants have been proposed for SPP. Each of the techniques offers a certain level of accuracy with particular limitations. The developments in machine learning processes fascinating the forecasting of stock market prices. Estimating direction of change (DOC) is one of the most important and hard issues in the stock price estimation process. This particular study examines how the temporal frame size might affect the success rate of SPP and DOC estimation with the time evolution. To do this, seven National Association of Securities Dealers Automated Quotations (NASDAQ) equities recorded between June 30, 2009, and April 3, 2024 were employed in the study. A MLP and an RNN with LSTM were used for prediction. The accuracy of the model was measured in accordance with the actual values. The error control was made with mean absolute error (MAPE). The results showed that there is an impact of frame size on the SPP as well as DOC estimation. Also, results showed that while LSTM performed better in long-term prediction, MLP performed better in shortterm prediction. © 2024 IEEE.
  • Küçük Resim Yok
    Öğe
    Predictive Modeling of Bacteria-Based Nanonetwork Performance Using Simulation-Driven Machine Learning and Genetic Algorithm Optimization
    (Wiley-V C H Verlag Gmbh, 2026) Duman, Mustafa Ozan; Isik, Ibrahim; Er, Mehmet Bilal; Tagluk, Mehmet Emin; Isik, Esme
    Bacteria-based nanonetwork (BN) offers a biologically inspired solution for enabling information exchange between nanomachines (NMs) in environments where traditional communication methods are ineffective. This study presents a 2D simulation model of a BN system that captures the chemotactic behavior of a single Escherichia coli (E. coli) bacterium navigating from a transmitter (TX) toward a receiver (RX) under varying environmental conditions. Key parameters, which are chemoattractant release rate (Q), TX-RX distance (d), and bacterial lifespan (), are systematically varied to evaluate their impact on communication performance, measured in terms of reach time and success rate. To enable accurate performance prediction without the need for computationally expensive repeated simulations, an analytical model is constructed using various machine learning (ML) techniques, including Linear Regression (LR), Random Forest (RF), and Multi-Layer Perceptron (MLP). Hyperparameters of MLP are optimized using a Genetic Algorithm (GA), significantly enhancing predictive accuracy and training stability. The results demonstrate the effectiveness of integrating dynamic simulation with data-driven modeling and hyperparameter optimization to represent complex system behavior. This framework offers valuable design insights for BN system development and supports the creation of efficient, scalable nanonetworks.
  • Küçük Resim Yok
    Öğe
    Yapay sinir ağı modelleri ile hisse senetlerinin değişim tahmininde farklı düğüm sayıları, zaman çerçeveleri ve aktivasyon fonksiyonlarının etkisinin araştırılması
    (İnönü Üniversitesi, 2025) Duman, Mustafa Ozan; Tağluk, Mehmet Emin
    Çok sayıda değişkenin etkisi altında olan sistematik değişkenlerin rastgele yürüyüşünü, modern tekniklerle dahi tahmin etmek oldukça zordur. Bu tür karmaşık problemlerin çözümünde, yapay zekâ tabanlı modellerin etkinliği ve hangi modelin hangi parametrelerle daha iyi sonuç vereceği araştırmacılar tarafından merak edilmektedir. Bu tezde, bu tür bir problem olan hisse senetlerinin gelecekteki değerlerini tahmin etmek için farklı YSA modelleri çeşitli parametrelerle test edilerek, bu parametrelerin tahmin performansı üzerindeki etkileri incelenmiştir. Çalışmada, BIST ve NASDAQ borsalarından seçilen bazı hisselere ait, farklı frekansta değişkenler içeren verilerle, farklı düğüm sayıları ve aktivasyon fonksiyonlarına sahip LSTM ve MLP modelleri çeşitli zaman çerçevelerinde test edilmiştir. Denemeler iki ana örnek üzerinden gerçekleştirilmiştir. İlk örnekte farklı düğüm sayıları ve aktivasyon fonksiyonlarının etkisi incelenmiştir. Bu kapsamda, ReLU veya sigmoid aktivasyon fonksiyonu kullanılan ve farklı sayılarda düğüm içeren MLP ile tahminler yapılmıştır. İkinci örnekte, çeşitli frekansta değişkenler içeren veriler kullanılarak, çeşitli modellerin ve farklı zaman çerçevelerinin tahmin sonuçları üzerindeki etkisi incelenmiştir. Burada 1, 2, 3, 4 ve 5 günlük zaman çerçevelerinde LSTM ve MLP ile tahminler yapılmıştır. Üretilen tahminler, gerçek değerlerle MAPE ve EDOC ile karşılaştırılmıştır. Sonuçlar, düğüm sayılarının yaklaşık hesap kuralıyla belirlenebileceğini ve aktivasyon fonksiyonu seçiminde ReLU'nun Sigmoid'e göre daha iyi bir seçenek olabileceğini göstermektedir. Daha fazla sayıda ve farklı frekansta değişken içeren verilerin kullanımının tahmin doğruluğunu artırdığı görülmüştür. Kısa zaman çerçevelerinde yapılan tahminlerin rastgele yürüyüş hipoteziyle uyumlu olarak başarız olduğu görülmüştür. Zaman çerçevelerinin uzunluğu arttıkça, MAPE'de çok büyük olmayan bir artışla birlikte yön doğruluğunun belirgin şekilde iyileştiği ve LSTM modelinin MLP'ye göre daha iyi sonuçlar verdiği görülmüştür. Bu durum, elde edilen sonuçların Dow teorisiyle uyumlu olduğunu ve hisse senetlerinin geçmiş değerlerinin, gelecekteki hareketleriyle bir ilişki taşıdığını göstermektedir.