Yazar "Er, Mehmet Bilal" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim
    Öğe
    Classification of thermoluminescence features of CaCO3 with long short-term memory model
    (Wiley, 2021) Işık, Esme; Toktamış, Dilek; Er, Mehmet Bilal; Hatib, Muhammed
    Calcium carbonate (CaCO3 ), a mineral commonly found in the Earth's crust, is mainly in the forms of calcite and aragonite. Calcite has the most stable type of thermodynamics at room temperature and ambient pressure. It has wide band gap structure and is of great interest for a wide-range technical and industrial applications due to its physical properties and suitability. In this study, a new method based on the long short-term memory (LSTM) model of deep learning is proposed to classify the thermoluminescence properties such as fading, cycle of measurement, heating rate, and dose-response of CaCO3 . Therefore the thermoluminescence properties of calcite was investigated as a suitable band structure and its coherent data were used to classify the features using a deep learning LSTM model. Experiments were carried out on a dataset consisting of four classes. The accuracy, precision, and sensitivity values of the proposed model obtained were 98.34, 97.90, and 98.56%, respectively. The classification process of the results obtained from the designed model showed good performance.
  • Küçük Resim
    Öğe
    Classification of thermoluminescence features of the natural halite with machine learning
    (TAYLOR & FRANCIS, 2022) Toktamış, Dilek; Er, Mehmet Bilal; Işık, Esme
    Radiation dosimeters are used to measure the absorbed radiation dose of any living organism during the time intervals. They include defective crystals that store radiation until they are stimulated. Thermoluminescence (TL) is a way to see the absorbed dose of the dosimeters. The irradiated crystal is heated up to 500°C to reveal the absorbed dose as a luminescence light. The TL dosimetric properties of natural halite (rock-salt) crystals extracted from Meke crater lake in Konya, Turkey, were investigated in this study. Support Vector Machine (SVM), Artificial Neural Network (ANN) and K-Nearest Neighbor (K-NN) were also examined utilizing machine learning for categorization of TL characteristics. According to the experimental output, the TL glow curve has two main peaks located at 100 and 270°C with good dosimetric properties. In the three classifiers, SVM has the biggest accuracy and precision. High training-low testing and results from normalized data give the best accuracy, precision, sensitivity and F-score.
  • Küçük Resim
    Öğe
    Parkinson's detection based on combined CNN and LSTM using enhanced speech signals with Variational mode decomposition
    (Elsevier, 2021) Er, Mehmet Bilal; Işık, Esme; Işık, İbrahim
    Parkinson's disease (PD) can cause many non-motor and motor symptoms such as speech and smell. One of the difficulties that Parkinson's patients can experience is a change in speech or speaking difficulties. Therefore, the right diagnosis in the early period is important in reducing the possible effects of speech disorders caused by the disease. Speech signal of Parkinson patients shows major differences compared to normal people. In this study, a new approach based on pre-trained deep networks and Long short-term memory (LSTM) by using mel-spectrograms obtained from denoised speech signals with Variational Mode Decomposition (VMD) for detecting PD from speech sounds is proposed. The proposed model consists of four steps. In the first step, the noise is removed by applying VMD to the signals. In the second step, mel-spectrograms are extracted from the enhanced sound signals with VMD. In the third step, pre-trained deep networks are preferred to extract deep features from the mel-spectrograms. For this purpose, ResNet-18, ResNet-50 and ResNet-101 models are used as pre-trained deep network architecture. In the last step, the classification process is occurred by giving these features as input to the LSTM model, which is designed to define sequential information from the extracted features. Experiments are performed with the PC-GITA dataset, which consists of two classes and is widely used in the literature. The results obtained from the proposed method are compared with the latest methods in the literature, it is seen that it has a better performance in terms of classification performance.