- Statistical Distribution Estimation and Applications
- Advanced Statistical Process Monitoring
- Target Tracking and Data Fusion in Sensor Networks
- Inertial Sensor and Navigation
- Advanced Statistical Methods and Models
- MRI in cancer diagnosis
- Advanced Adaptive Filtering Techniques
- Face and Expression Recognition
- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Statistical Methods and Inference
- Reliability and Maintenance Optimization
- Geophysical and Geoelectrical Methods
- High voltage insulation and dielectric phenomena
- Probabilistic and Robust Engineering Design
- Smart Parking Systems Research
- Fault Detection and Control Systems
- Advanced Optimization Algorithms Research
- Stochastic Gradient Optimization Techniques
- Cultural and Sociopolitical Studies
- Currency Recognition and Detection
- Control Systems and Identification
- Hydrology and Drought Analysis
- Sparse and Compressive Sensing Techniques
- Advanced Numerical Analysis Techniques
Kırıkkale University
2016-2024
Ankara University
2012
In the vast statistical literature, there are numerous probability distribution models that can model data from real-world phenomena. New models, nevertheless, still required in order to represent with various spread behaviors. It is a known fact great need for new limited support. this study, flexible called unit Maxwell-Boltzmann distribution, which values interval, derived by selecting as base-line model. The important characteristics of terms statistics and mathematics investigated...
The geometric process (GP) is a simple and direct approach to modeling of the successive inter-arrival time data set with monotonic trend. In addition, it quite important alternative non-homogeneous Poisson process. present paper, parameter estimation problem for GP considered, when distribution first occurrence Power Lindley parameters α λ . To overcome GP, maximum likelihood, modified moments, L-moments least-squares estimators are obtained a, mean, bias mean squared error (MSE) values...
The aim of this study is to investigate the solution statistical inference problem for geometric process (GP) when distribution first occurrence time assumed be Rayleigh. Maximum likelihood (ML) estimators parameters GP, where a and λ are ratio parameter GP scale Rayleigh distribution, respectively, obtained. In addition, we derive some important asymptotic properties these such as normality consistency. Then run simulation studies by different values compare estimation performances obtained...
We introduce a new wrapped exponential distribution named transmuted (TWE) distribution, for the modeling of circular datasets by using Transmutation Rank-Map method. This method is employed first time with this study. The introduced more flexible than traditional distribution. paper provides explicit form important distributional properties such as expectation, median, moments, characteristic function, quantile hazard rate function and stress-strength reliability. Rényi Shannon entropies...
Breast cancer is the most common that progresses from cells in breast tissue among women. Early-stage detection could reduce death rates significantly, and detection-stage determines treatment process. Mammography utilized to discover at an early stage prior any physical sign. However, mammography might return false-negative, which case, if it suspected lesions have of chance greater than two percent, a biopsy recommended. About 30 percent biopsies result malignancy means rate unnecessary...
The extended Kalman filter is extensively used in nonlinear state estimation problems. As long as the system characteristics are correctly known, gives best performance. However, when information partially known or incorrect, may diverge give biased estimates. An extensive number of works has been published to improve performance filter. Many researchers have proposed introduction a forgetting factor, both into and filter, there 2 fundamental problems with this approach: incorporation...
The problem of statistical modeling the geometric count data with a specific probability model lifetimes is interest and importance in reliability. In this paper, we construct process (GP), parameter [Formula: see text], for when distribution first occurrence time scaled Muth parameters text] text]. We investigate estimators from point approximations classical modified approach by using different estimation methodologies such as maximum likelihood, moments, least-squares spacing. perform...
Abstract The geometric process is one of the important simple monotonic processes with a positive ratio parameter in theory stochastic processes. Simply, it can be thought as generalization renewal (RP). In current paper, we mainly study Hjorth marginal distribution, parameters θ and λ, for being able to model successive inter‐arrival times trend. We first examine inference problem mentioned from different perspectives obtain estimators its by employing estimation methods such maximum...
Uygulamalarda gözlemlenen verilerin en uygun biçimde istatistiksel analizini yapmak için veri kümesinin altında yatan dağılım belirlenmelidir. Çoğu zaman, bir dağılımı belirlemeye çalışırken kullanılan uyum iyiliği testleri, seti birden fazla modelini işaret eder. Uyum testlerinin sonuçlarına göre olası modelleri arasında, kümesi optimal modelinin belirlenmesi problemi, istatistikte oldukça önemli problemdir. Bu çalışmada, geometric süreç verileri Gamma ve Weibull dağılımları arasındaki...
Abstract In this paper, the stability of adaptive fading extended Kalman filter with matrix forgetting factor when applied to state estimation problem noise terms in non–linear discrete–time stochastic systems has been analysed. The analysis is conducted a similar manner standard filter’s based on framework. theoretical results show that under certain conditions initial error and terms, remains bounded stable. importance contribution performance adaptation method are demonstrated...
In the present paper, statistical inference problem is considered for geometric process (GP) by assuming distribution of first arrival time generalized Rayleigh with parameters $\alpha$ and $\lambda$. We use maximum likelihood method obtaining ratio parameter GP distributional distribution. By a series Monte-Carlo simulations evaluated through different samples sizes small, moderate large, we also compare estimation performances estimators other available in literature such as modified...
Gamma ve Weibull dağılımları sağlık, güvenilirlik, mühendislik vb.ortak uygulama alanlarına sahip olan dağılımlardır.Çoğu zaman bu iki dağılım bir veri seti için benzer sonuç çıkarımlar sağlasa da (çakışsa da), setini en iyi modelleyecek dağılımın seçilmesi arzulanır.Bu çalışmada, ya dağılımlarından herhangi birinden gözlendiği
Heterojen yapıda bir popülasyondan elde edilmiş verilerin istatistiksel analizinde oldukça kullanışlı modeller olan sonlu karma dağılımlar için parametre tahmin problemi istatistikte önemli problemdir. Bu çalışma, iki parametreli Rayleigh dağılımlarının karmaları problemini ele almaktadır. kapsamda, karmalarında mevcut bilinmeyen parametreler en çok olabilirlik edicileri E-M algoritmasına göre edilmektedir. Bununla birlikte çalışmada, edilen edicilerinin dağılımın parametrelerini etmedeki...
Discriminant analysis is defined as a statistical technique that classifies unit whose properties are measured, into one of the known finite numbers populations. In this classifying process, an error occurs when classified to different population from its own population. This called rate or probability incorrect classification. It desirable minimize error. study focuses on determining parameter estimation method provides minimum rate, parameters Weibull populations not known. Maximum...
Low-rank matrix approximations have recently gained broad popularity in scientific computing areas. They are used to extract correlations and remove noise from matrix-structured data with limited loss of information. Truncated singular value decomposition (SVD) is the main tool for low-rank approximation. However, applications such as latent semantic indexing where document collections dynamic over time, i.e. term subject repeated updates, SVD becomes prohibitive due high computational...