Yazar "Elen, Abdullah" seçeneğine göre listele
Listeleniyor 1 - 9 / 9
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe A comparison of classification methods for diagnosis of Parkinson's(İsmail Sarıtaş, 2020) Elen, Abdullah; Avuçlu, EmreParkinson's is a neurological health problem and one of the most common diseases affecting more than four million people worldwide. Recent studies have shown that deterioration of vocal cords, especially from Parkinson's, provides important information in the diagnosis and follow-up of the disease. In this study, a database of biomedical voice recordings from 32 people of different ages and genders was used to diagnose Parkinson's disease. With this database, the performance comparison of the machine learning algorithms k-Nearest Neighborhood (k-NN) and Naïve Bayes (NB) classifiers were performed. Seven different distance measurement methods (Chebychev, Correlation, Cosine, Euclidean, Hamming, Mahalanobis, and Spearman) for the k-NN and five different distribution methods (Uniform kernel, Epanechnikov kernel, Gaussian kernel, Triangular kernel and Normal distribution) for the NB classifier were performed in the performance process and separate tests were performed. The data obtained from these tests were compared with statistical measurements. In experimental studies, we used 10-fold cross validation technique for Parkinson dataset. Better results were obtained from k-NN classification algorithm than Naive Bayes classification algorithm. While k-NN mean accuracy score was 82.34%, this ratio was obtained as 74.15% for NB. Mahalanobis distance measurement method was found to give better results.Öğe A hybrid machine learning model for classifying time series(Springer Science and Business Media Deutschland GmbH, 2022) Elen, Abdullah; Avuçlu, EmreA time series is a sequence of numerical data points in equal time intervals and/or successive order. Time series are used in many fields to understand the behavior of systems, to make predictions, to create solutions to problems, etc. Electroencephalogram (EEG) and electrocardiogram (ECG) are frequently used in the diagnosis of diseases and research in this field. EEG signals examine the neural activity of the brain, while ECG examines the work of the heart muscle and neural conduction system. These signs contain a large amount of information about the functioning of the brain and heart functions. In order to use this information, experts in the field of signal processing must evaluate these signals. Due to the successful application methods of EEG and ECG signals in classification problems, various fields of artificial intelligence applications are frequently used by experts. In this study, a new hybrid model has been developed to classify EEG and ECG signals. These signals of five different classes have been used as the feature vector in the training of machine learning algorithms with 10 statistical parameters (8 normalized, 2 real signals). These algorithms are designed to give the best performance. In the proposed hybrid model, a machine learning model consisting of four stages is used. In the experimental studies, it has been observed that the proposed hybrid method gives better results than the normal classification process. The obtained results are given in the experimental studies section in detail.Öğe A new dynamic feature extraction method for biometric images(Gazi Üniversitesi, 2021) Avuçlu, Emre; Elen, Abdullah; Özçifçi, AyhanThe image of biometric properties in humans is used in many fields today. Regardless of these features, it is necessary to first translate it into data that the computer understands. In this study, automatic and dynamic image segmentation was performed by using 300x300 fingerprint images. A fingerprint database with a total of 80 images and 10 different classes was used. The features of the images were subtracted from the sub-segments obtained from these images by the feature extraction algorithm that was originally developed. The 300x300 images were divided into 25x25 sub-images and the feature vector was obtained. 144x80 inputs obtained after image segmentation were kept in areas in separate tables. The developed segmentation and feature extraction algorithm can be applied to any image of equal size.Öğe An application to control media player with voice commands(Gazi Üniversitesi, 2020) Avuçlu, Emre; Özcifçi, Ayhan; Elen, AbdullahUsing technology today is of great importance in terms of making people's lives easier. It has become very easy to run some applications with technology. In this study, an application that provides media player control with voice commands was developed. This application was developed to address the needs of people who cannot listen to music on their own due to any disability. The application was implemented in C# programming language. In order to manage the media player with voice commands, voice recognition libraries were first used. In the developed application, operations with keyboard and mouse can be done with voice commands. Voice commands can be sent with the wireless headset from anywhere in the shooting area.Öğe Automatic detection of petiole border in plant leaves(SAGE Publications Ltd, 2021) Elen, Abdullah; Avuçlu, EmrePlants are our source of oxygen and nutrients on earth. Therefore, conservation of biodiversity is vital for the survival of other species. With the developing technology, plant species can be examined more closely. Image processing, which is a subject of computer science, has an important role in this field. In this study, an image processing–based method has been developed to automatically separate the petiole region of the plant leaves. To determine the boundary line of the petiole region, the cumulative pixel distributions of the input images in binary format according to the X- and Y-axis are analyzed. Accordingly, optimum thresholds and petiole boundary points are determined. The proposed method was tested on 795 leaf images from 90 different plant species that grow both as trees and shrubs in the Czech Republic. According to the results obtained in experimental studies, it is thought that the proposed method will make an important contribution especially in studies such as automatic classification of plants and leaves and determination of plant species in botanical science.Öğe Designing a dynamic system to control building components and computers with voice commands for disabled individuals(Springer, 2023) Avuçlu, Emre; Özçifçi, Ayhan; Elen, AbdullahNowadays, with the development of technology, there have been many innovations in human life. Technology has made many things easier in human life depending on these innovations. In this study, an application has been developed to more easily solve the needs of people in their daily lives. The developed application is designed to control the internal units of the house with voice commands. The doors, windows, lamps, etc. that we use daily in our house can be controlled by voice commands through wireless headphones. Voice commands can be sent from inside or outside the house. In addition to these external units, all applications on the personal computer are controlled by voice commands without using a keyboard and mouse. The application developed in C # programming language, provides control of newly installed programs to the computer without coding. To do so, it is enough to add that program and voice command to the interface. Moreover, the application provides a lot of convenience for disabled-aged citizens and some dangerous situations in the industrial environment.Öğe Evaluation of train and test performance of machine learning algorithms and Parkinson diagnosis with statistical measurements(Springer Science and Business Media Deutschland GmbH, 2020) Avuçlu, Emre; Elen, AbdullahParkinson’s disease is a neurological disorder that causes partial or complete loss of motor reflexes and speech and affects thinking, behavior, and other vital functions affecting the nervous system. Parkinson’s disease causes impaired speech and motor abilities (writing, balance, etc.) in about 90% of patients and is often seen in older people. Some signs (deterioration of vocal cords) in medical voice recordings from Parkinson’s patients are used to diagnose this disease. The database used in this study contains biomedical speech voice from 31 people of different age and sex related to this disease. The performance comparison of the machine learning algorithms k-Nearest Neighborhood (k-NN), Random Forest, Naive Bayes, and Support Vector Machine classifiers was performed with the used database. Moreover, the best classifier was determined for the diagnosis of Parkinson’s disease. Eleven different training and test data (45 × 55, 50 × 50, 55 × 45, 60 × 40, 65 × 35, 70 × 30, 75 × 25, 80 × 20, 85 × 15, 90 × 10, 95 × 5) were processed separately. The data obtained from these training and tests were compared with statistical measurements. The training results of the k-NN classification algorithm were generally 100% successful. The best test result was obtained from Random Forest classifier with 85.81%. All statistical results and measured values are given in detail in the experimental studies section.Öğe Making inferences about settlements from satellite images using glowworm swarm optimization(Korean Institute of Electrical Engineers, 2020) Avuçlu, Emre; Elen, Abdullah; Kahramanlı Örnek, HumarOptimization is the process of choosing the best one among existing possibilities under particular circumstances in a problem. There are various algorithms for optimization problems nowadays. Metaheuristic algorithms are the algorithms giving almost optimum solutions at an acceptable duration for the problems of large dimension. Heuristic optimization algorithms with general aim are evaluated in different groups. Swarm intelligence-based optimization algorithms were developed through examining the behaviors and movements of living flocks such as birds, fish, cats, and bees. With these algorithms, some estimating processes are carried out successfully in all areas. In this study a new approach was presented with a novel idea, by inspiring from the behavior type of Glowworm Swarm Optimization; and an application estimating the total population, square measurement and electricity quantity that was consumed by the chosen areas in a region was developed. The developed application works as a real-time and animated display. When all calculations are finished, the animation ends. Estimates also examined England as an example. The difference between the estimated value of the actual population of England is calculated as 1.7%. In the estimates for the values of the surface area of England with an error of 1.4%, the estimated values were very close to the actual values. Some other obtained estimation results are presented in the results section.Öğe Standardized Variable Distances: A distance-based machine learning method(Elsevier, 2021) Elen, Abdullah; Avuçlu, EmreToday, machine learning algorithms are an important research area capable of analyzing and modeling data in any field. Information obtained through machine learning methods helps researchers and planners to understand and review systematic problems of their current strategies. Thus, it is very important to work fully in every field that facilitates human life, such as early and correct diagnosis, correct choice, fully functioning autonomous systems. In this paper, a novel machine learning algorithm for multiclass classification is presented. The proposed method is designed based on the Minimum Distance Classifier (MDC) algorithm. The MDC is variance-insensitive because it classifies input vectors by calculating their distances/similarities with respect to class-centroids (average value of input vectors of a class). As it is known, real-world data contains certain proportions of noise. This situation negatively affects the performance of the MDC. To overcome this problem, we developed a variance-sensitive model, which we call Standardized Variable Distances (SVD), considering the standard deviation and z-score (standardized variable) factors. To ensure the accuracy of the SVD, we used Wisconsin Breast Cancer Original (WBCO) and LED Display Domain (led7digit) datasets, which we obtained from UCI machine learning repository, with 5-fold cross validation. It was compared and analyzed classification performance of the SVD with Decision Tree (DT), Random Forest (RF), k-Nearest Neighbor (k-NN), Multinomial Logistic Regression (MLR), Naive Bayes (NB), Support Vector Machine (SVM), and the Minimum Distance Classifier (MDC), which are well-known in the literature. It has also been compared thirteen different studies using the same datasets over the past five years. Our results in the experimental studies have shown that the SVD can classify better than traditional and state-of-the-art methods, compared in this study. The proposed method reached over 97% classification accuracy (CACC), F-measure (FM) and area under the curve (AUC) on the WBCO dataset. On the led7digit dataset, approximately 74% CACC, 75.1% FM and 82.2% AUC scores were obtained. It has been observed that the classification scores obtained with the SVD are higher than other ML algorithms used in the experimental studies.
