Sürdürülebilir Kalkınma Amaçları
Yayınlar
- MakaleYalnız künye2022Detection of damaged buildings after an earthquake with convolutional neural networks in conjunction with image segmentation
Ünlü, Ramazan; Kiriş, Recep
Detecting damaged buildings after an earthquake as quickly as possible is important for emergency teams to reach these buildings and save the lives of many people. Today, damaged buildings after the earthquake are carried out by the survivors contacting the authorities or using some air vehicles such as helicopters. In this study, AI-based systems were tested to detect damaged or destroyed buildings by integrating into street camera systems after unexpected disasters. For this purpose, we have used VGG-16, VGG-19, and NASNet convolutional neural network models which are often used for image recognition problems in the literature to detect damaged buildings. In order to effectively implement these models, we have first segmented all the images with the K-means clustering algorithm. Thereafter, for the first phase of this study, segmented images labeled “damaged buildings” and “normal” were classified and the VGG-19 model was the most successful model with a 90% accuracy in the test set. Besides, as the second phase of the study, we have created a multiclass classification problem by labeling segmented images as “damaged buildings,” “less damaged buildings,” and “normal.” The same three architectures are used to achieve the most accurate classification results on the test set. VGG-19 and VGG-16, and NASNet have achieved considerable success in the test set with about 70%, 67%, and 62% accuracy, respectively.
- MakaleYalnız künye2021A reduced variance unsupervised ensemble learning algorithm based on modern portfolio theory
Ünlü, Ramazan; Xanthopoulos, Petros
Unsupervised ensemble learning or consensus clustering has gained popularity due to its ability to combine multiple clustering solutions into a single solution that is robust and often performs better than the individual ones. There have been several approaches to consensus clustering including voting and weighted voting algorithmic schemes. Although there have been several algorithms for adjusting the weights of a consensus clustering all of them are tuned based on some performance characteristic associated with clustering accuracy. In this paper, we propose a method for incorporating weights by taking into consideration the intra algorithmic variability i.e. algorithms that provide solutions with very different performance upon multiple runs. The methodology is inspired by modern portfolio theory and more specifically from Markowitz model for asset allocation where one is trying to identify the most efficient portfolio through the solution of a convex optimization problem. Here, efficiency is defined as the minimum amount of risk for an expected return. We apply this method to different datasets and compare with respect to performance and robustness. The proposed scheme appears to achieve competitive average performance with very low variability.
- Kitap BölümüYalnız künye2021Safer-Driving: Application of Deep Transfer Learning to Build Intelligent Transportation Systems
Ünlü, Ramazan
Deep learning methods are widely used in the field of transportation as well as in many areas. These days, deep learning methods produce very successful results thanks to big data, where data collection and storage are much easier than in past years. However, it can be quite difficult to collect data or find the right data in some cases. On the other hand, even if the desired data is collected enough, training deep neural networks to learn from these data may not be comfortable in terms of hardware needed for ordinary users. In these cases, transfer learning can be an exit point, which is very popular in deep understanding, especially when the amount of data is limited. In this study, we have aimed to give comprehensive guidance on using a transfer learning strategy to create a detection model to establish a safer transportation system. More specifically, we have developed a classification model to alleviate accidents due to deer on the road. We have implemented VGG-16 convolutional neural network architecture as a part of the structured transfer learning framework. Computational results show that high accuracy performance can be achieved by implementing VGG-16 with the respectively small size of a dataset.
- Kitap BölümüYalnız künye2021An Assessment of Imbalanced Control Chart Pattern Recognition by Artificial Neural Networks
Ünlü, Ramazan
Manual detection of abnormality in control data is an annoying work which requires a specialized person. Automatic detection might be simpler and effective. Various methodologies such as ANN, SVM, Fuzzy Logic, etc. have been implemented into the control chart patterns to detect abnormal patterns in real time. In general, control chart data is imbalanced, meaning the rate of minority class (abnormal pattern) is much lower than the rate of normal class (normal pattern). To take this fact into consideration, authors implemented a weighting strategy in conjunction with ANN and investigated the performance of weighted ANN for several abnormal patterns, then compared its performance with regular ANN. This comparison is also made under different conditions, for example, abnormal and normal patterns are separable, partially separable, inseparable and the length of data is fixed as being 10, 20, and 30 for each. Based on numerical results, weighting policy can better predict in some of the cases in terms of classifying samples belonging to minority class to the correct class.
- MakaleYalnız künye2021A robust data simulation technique to improve early detection performance of a classifier in control chart pattern recognition systems
Ünlü, Ramazan
The quality control process is essential in maintaining the stability of production systems and proactively detecting abnormalities that may result in high mechanical and labor costs. In this study, a new data simulation strategy was devised to improve the early prediction performance (EPP) of an algorithm. Traditional control chart data simulation methods work based on creating abnormal patterns consisting of only abnormal signals. However, a classifier trained with data samples consisting of only abnormal data signals may fail to early detect abnormality in a real-time production line, in which abnormal signals is obscured by volume of normal signals. From this perspective, training a model by imitating real-world cases can improve the performance of an algorithm in terms of early detection of an abnormality. Normal and abnormal patterns were simulated by implementing a new approach called Mix Ratio Data Simulation (MRDS). The proposed methodology MRDS is compared with the customary data simulation method under the predefined scenarios in terms of EPP. The findings indicated that changing the way of simulating dataset increases the EEP of the machine-learning algorithm regardless of abnormality types and parameters.
- MakaleYalnız künye2021Cost-oriented LSTM methods for possible expansion of control charting signals
Ünlü, Ramazan
Manual quality control may result in delayed detection of a system defect, or none at all, potentially resulting in malfunctions that can lead to disruption of the system, incur extra costs, or complete downtime of the system. To mitigate such issues, more advanced statistics and machine learning-based systems such as SVM, WSVM, etc. are used to automatically detect error signals during the process. However, these methodologies are not developed for sequential dataset such as a classical CCPR dataset. As a result, in this study, we have implemented a preliminary Cost-Oriented Long-Short Term Memory (LSTM), which is designed to learn from a sequential dataset and compare it with SVM and WSVM which are traditional methods utilized in the field. Additionally, we compared the performance of methods in terms of early detection of an abnormal pattern. Based on the findings, the Cost-Oriented LSTM method outperforms SVM and WSVM in the majority of abnormal patterns in terms of both classification and early detection performance. Global accuracy score which is the average of accuracy rates in all combination of abnormal parameters and window lengths shows that LSTM gives a better accuracy score than SVM and WSVM in all seven abnormal patterns detection.
- MakaleAçık Erişim2020Verimlilikte Yapay Zeka’nın Rolü: Şarap Kalitesinin Tahminine Yönelik Bir Vaka Çalışması
Ünlü, Ramazan
Yapay zeka son yıllarda birçok alanda kullanılmaya başlanmış ve oldukça başarılı sonuçlar elde edilmiştir. Sağlık sektöründensürücüsüz araçlara kadar birçok alanda kullanılan yapay zeka, üretim sektöründe de verimliliğin artırılması için sıklıkla kullanılmıştır.Bu çalışmada normal olarak dağılmamış bir veri setinde yapay zeka algoritmalarının kullanılmasına yönelik bir çerçeve çizilmeyeçalışılmıştır. Normal dağılım gösteren veri setlerinde herhangi bir yapay zeka algoritması kolaylıkla uygulanabilirken normal dağılımgöstermeyen veri setlerinde ya verinin kendisine farklı bir işlem uygulanması gerekir veya algoritmanın teorik yapısının revize edilmesigerekmektedir. Bu açıdan bu çalışmada iki farklı yöntemde uygulanmıştır. İlk olarak literatürde sıklıkla kullanılan Destek VektörMakinaları kullanılmıştır. Buna ek olarak Destek Vektör Makinalarının normal dağılmayan veri setlerinde başarılı sonuçlar vermesiiçin uyarlanmış şekli olan Ağırlıklandırılmış Destek Vektör Makineleri uygulanmıştır. Son olarak Sentetik Azınlık Aşırı ÖrneklemeTekniği (SMOTE) tekniği uygulanmış ve kullanılan veri seti yapay olarak normal dağılıma yakınsanmıştır. Kullanılan üç teknikteduyarlılık, hassaslık, özgüllük, yaygınlık, F skor ve Geometrik Ortalama (G-Mean) değerlendirme kriterleri açısından karşılaştırılmıştır.Çalışma sonucuna göre Ağırlıklandırılmış Destek Vektör Makineleri kullanılan değerlendirme kriterlerine göre en başarılı sonuçlarıvermiştir.
- MakaleYalnız künye2020An assessment of machine learning models for slump flow and examining redundant features
Ünlü, Ramazan
Over the years, several machine learning approaches have been proposed and utilized to create a prediction model for the high-performance concrete (HPC) slump flow. Despite HPC is a highly complex material, predicting its pattern is a rather ambitious process. Hence, choosing and applying the correct method remain a crucial task. Like some other problems, prediction of HPC slump flow suffers from abnormal attributes which might both have an influence on prediction accuracy and increases variance. In recent years, different studies are proposed to optimize the prediction accuracy for HPC slump flow. However, more state-of-the-art regression algorithms can be implemented to create a better model. This study focuses on several methods with different mathematical backgrounds to get the best possible results. Four well-known algorithms Support Vector Regression, M5P Trees, Random Forest, and MLPReg are implemented with optimum parameters as base learners. Also, redundant features are examined to better understand both how ingredients influence on prediction models and whether possible to achieve acceptable results with a few components. Based on the findings, the MLPReg algorithm with optimum parameters gives better results than others in terms of commonly used statistical error evaluation metrics. Besides, chosen algorithms can give rather accurate results using just a few attributes of a slump flow dataset.
- MakaleAçık Erişim2020Machine learning and classical forecasting methods based decision support systems for covid-19
Ünlü, Ramazan; Namli, Ersin
From late 2019 to the present day, the coronavirus outbreak tragically affected the whole world and killed tens of thousands of people. Many countries have taken very stringent measures to alleviate the effects of the coronavirus disease 2019 (COVID-19) and are still being implemented. In this study, various machine learning techniques are implemented to predict possible confirmed cases and mortality numbers for the future. According to these models, we have tried to shed light on the future in terms of possible measures to be taken or updating the current measures. Support Vector Machines (SVM), Holt-Winters, Prophet, and Long-Short Term Memory (LSTM) forecasting models are applied to the novel COVID-19 dataset. According to the results, the Prophet model gives the lowest Root Mean Squared Error (RMSE) score compared to the other three models. Besides, according to this model, a projection for the future COVID-19 predictions of Turkey has been drawn and aimed to shape the current measures against the coronavirus.
- MakaleAçık Erişim2019Classification of Historical Anatolian Coins with Machine Learning Algorithms
Ünlü, Ramazan
To find out which period the historical coins belong to requires a number of scientific procedures that archaeologists or expertscan do. These operations can often be time-consuming and demanding operations. From this point on, in this study, theautomatically classification of historical coins by using machine learning methods is discussed. Being able to use machine learningmethods to classify historical coins can help experts and can become an analysis tool without the need for scientific tests fornon-experts. For this purpose, some physical properties of different coins used in Anatolian geography were collected andclassified by various machine learning methods named SVM, Random Forest, Bagging, and Decision Trees. Also, two differentmissing values strategies are deployed in conjunction with each chosen method. Based on our findings, random forest methodtogether with imputing missing values with mean gives an acceptable results with the accuracy rate of %71, although there aresome limitations such as high rate of missing values and working with a small dataset
- Kitap BölümüYalnız künye2019An assessment of imbalanced control chart pattern recognition by artificial neural networks
Ünlü, Ramazan
Manual detection of abnormality in control data is an annoying work which requires a specialized person. Automatic detection might be simpler and effective. Various methodologies such as ANN, SVM, Fuzzy Logic, etc. have been implemented into the control chart patterns to detect abnormal patterns in real time. In general, control chart data is imbalanced, meaning the rate of minority class (abnormal pattern) is much lower than the rate of normal class (normal pattern). To take this fact into consideration, authors implemented a weighting strategy in conjunction with ANN and investigated the performance of weighted ANN for several abnormal patterns, then compared its performance with regular ANN. This comparison is also made under different conditions, for example, abnormal and normal patterns are separable, partially separable, inseparable and the length of data is fixed as being 10,20, and 30 for each. Based on numerical results, weighting policy can better predict in some of the cases in terms of classifying samples belonging to minority class to the correct class.
- MakaleYalnız künye2019Estimating the number of clusters in a dataset via consensus clustering
Ünlü, Ramazan; Xanthopoulos, Petros
In unsupervised learning, the problem of finding the appropriate number of clusters -usually notated as k- is very challenging. Its importance lies in the fact that k is a vital hyperparameter for the most clustering algorithms. One algorithmic approach for tacking this problem is to apply a certain clustering algorithm with various cluster configurations and decide to use the one that maximizes a certain internal validity measure. This is a promising and computationally efficient approach since the independent runs are parallelizable. In this paper, we attempt to improve over this estimation approach by incorporating a consensus clustering approach into k estimating scheme. The weighted consensus clustering scheme employs four different indices namely Silhouette (SH), Calinski–Harabasz (CH), Davies–Bouldin (DB), and Consensus (CI) indices to estimate the correct number of cluster. Computational experiments in a dataset with clusters ranging from 2 to 7 show the profound advantages of weighted consensus clustering for correctly finding k in comparison to individual clustering method (e.g, k-means) and simple consensus clustering.