Yazar "Basaran, Murat Alper" seçeneğine göre listele

Listeleniyor 1 - 5 / 5
  • [ X ]
    Öğe
    AN INNOVATIVE HYBRID APPROACH TO FORECASTING INTERVAL TIME SERIES DATA WITH ELMAN ARTIFICIAL NEURAL NETWORKS AND A MODIFIED ADAPTIVE NETWORK- BASED FUZZY INFERENCE SYSTEM
    (2025) İslamoğlu, Ebrucan; Basaran, Murat Alper
    Interval Time Series (ITS) techniques are used in data analysis for modeling and forecasting. This paper introduces a hybrid model that combines two effective forecasting methods: the Modified Adaptive Network-Based Fuzzy Inference System (MANFIS) and the Elman Recurrent Neural Network (ERNN). Unlike non-interval time series data, ITS considers both the highest and lowest values within an interval to represent dynamic data, thereby capturing potential relationships between these bounds. The proposed algorithm incorporates ANFIS and ERNN structures with the following advantages: it utilizes particle swarm optimization to train the model and addresses both linear and nonlinear forecasting aspects, providing model-based and data-based approaches, respectively.The fuzzification process employs the fuzzy c-means clustering technique to systematically derive membership values from input data, thereby enhancing forecasting accuracy. The proposed method has been rigorously validated using seven diverse real-world datasets, and comparative analyses with existing algorithms in the literature confirm its superior performance.
  • [ X ]
    Öğe
    Determining factors affecting healthcare service satisfaction utilizing fuzzy rule-based systems
    (Taylor & Francis Ltd, 2016) Demir, Mehmet Ozer; Basaran, Murat Alper; Simonetti, Biagio
    Health communication, which is a multi-attribute concept, is a generic title describing clinical practice. The literature shows that the relation between health communication and healthcare service satisfaction (HSS) has been found to be significant. The main objective of pursuing better health communication is to achieve the best outcome and patient satisfaction where healthcare systems are supposed to deliver. However, the health communication is a complex process. Also, measuring patients' satisfaction is not an easy task since satisfaction is a complex notion with several factors. In this study, questions in the questionnaire directed to patients are factor-analyzed in order to obtain components which are used as independent attributes that will be modeled by fuzzy rule-based systems (FRBS) in order to explain HSS. Utilizing FRBS brings two different advantages, one of which is to use mathematical functions called membership functions for linguistically expressed responses. The second one is to observe the transition among the linguistic values expressed by patients. The four independent variables, namely, doctor-patient communication (DPC), information seeking behavior (ISB), equal behavior and tolerance to cultural differences (TCD) and the dependent variable HSS are employed in the modeling. Although both DPC and ISB have positive effects on HSS, TCD has none. One interesting finding about DPC is that if DPC scores below the average value tend to lower, it does not have a decreasing effect on HSS, which means that if a patient does expect to have average DPC, his or her evaluation on HSS does not lower, which says that if a patient knows that the doctor has a poor communication skill, the patient does not pay attention to this attribute.
  • [ X ]
    Öğe
    Dynamic response optimization of a thermoplastic composite sandwich beam under random vibration
    (Taylor & Francis Inc, 2024) Oktav, Akin; Basaran, Murat Alper; Daricik, Fatih
    The dynamic response of a thermoplastic composite sandwich structure is optimized under random vibration. First, the experimental modal analysis data of a set of test samples are processed by a sequential set of statistical analysis such as descriptive statistics, factor analysis, and paired sample t-test. Then, the sample with the highest ability to represent the group is taken as the reference data. Three different computational models, which are defined according to whether the solid to be meshed is considered an area or a volume, are constructed. Modal analysis results of the computational models are compared to the reference experimental data to evaluate the performance of the models. To predict the dynamic response of the sandwich beam, it is excited through a random signal in the transverse direction. The nodal acceleration responses are computed in 17 evenly spaced points located on the upper finishing layer of the sandwich beam. Finally, a geometry optimization study is conducted to predict the optimum thicknesses of the 7 layers bonded together to form the sandwich beam. The optimum layer thicknesses that minimize the nodal accelerations at 17 evenly spaced points on the sandwich beam are computed. The current study shows that the shell model has the closest values to the experimental data compared to other models. As far as the dynamic response of a TPC sandwich structure is concerned, it is concluded that the shell model better represents the structure during the modeling phase and leads to concurrently reduced weight and nodal acceleration, when optimized.
  • [ X ]
    Öğe
    Evaluation of Dynamic Response Variability in Aluminum Honeycomb Sandwich Panels Using PCE and Kriging-Based Metamodel
    (Gazi Univ, 2025) Oktav, Akin; Basaran, Murat Alper
    The aim of this study is to model the dynamic response variability in commercial aluminum honeycomb sandwich panels. An experimental modal analysis is performed on 35 identical commercial AHSPs. Based on 10,000 samples, a computational model is constructed according to the estimated weight of the panel. The modal frequencies of the 35 samples for the first 10 flexible modes are compared with the computational results and deviations are referred to as errors. The thickness of the facing sheets and thickness of the cell wall of the core are considered as sources of uncertainty. A data-driven meta-model called PCE-Kriging is created.
  • [ X ]
    Öğe
    Konaklama İşletmelerinin Tavsiye Edilmesini Etkileyen Hizmetlerin Belirlenmesi: Alanya’da Bir Araştırma
    (2020) Dogan, Seden; Basaran, Murat Alper; Kantarcı, Kemal
    İnternet üzerindeki çeşitli platformlarda yer alan kullanıcı değerlendirme ve puanlamalarının, diğer kullanıcıların seyahat kararı üzerindeki önemini ve etkisini ortaya koyan pek çok araştırma vardır. Söz konusu değerlendirme ve puanlamaların, konaklama işletmesinin tavsiye edilmesi üzerindeki etkisini inceleyen araştırmalar son birkaç yıldır yapılmakta olup sayısı nispeten azdır. Bu doğrultuda araştırmanın amacı HolidayCheck.de web sitesinde Alanya’da faaliyet gösteren dört ve beş yıldızlı konaklama işletmeleri için 2009-2017 yılları arasında farklı kriterler için verilen puanların, işletmelerin tavsiye edilmesi üzerindeki etki düzeyinin veen etkili değişkenin belirlenmesidir. HolidayCheck.de web sitesinde genel durum, lokasyon, çalışanlar, yiyecek-içecek hizmetleri, spor-havuz-eğlence hizmetleri ve odalar olarak belirlenen altı kritere verilen puanlamalara İkili Lojistik Regresyon Analizi uygulanmıştır. Elde edilen bulgulara göre konaklama işletmelerinin tavsiye edilmesi üzerinden en yüksek etkiye sahip kriter yiyecek-içecek hizmetlerive en düşük etkiye sahip kriter odalar olarak belirlenmiştir. İkinci derecede etkilikriter iseaile dostu, engelli dostu gibi özelliklerin puanlandığı genel durum olarak tespit edilmiştir.