16 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 10 / 16
Öğe A novel approach for spam email detection based on shifted binary patterns(Wiley-Blackwell, 2016-01-11) Kaya, Yılmaz; Ertuğrul, Ömer FarukAdvances in communication allow people flexibility to communicate in various ways. Electronic mail (email) is one of the most used communication methods for personal or business purposes. However, it brings one of the most tackling issues, called spam email, which also raises concerns about data safety. Thus, the requirement of detecting spams is crucial for keeping the users safe and saving them from the waste of time while tackling those issues. In this study, an effective approach based on the probability of the usage of the characters that has similar orders with respect to their UTF-8 value by employing shifted one-dimensional local binary pattern (shifted-1D-LBP) was used to extract quantitative features from emails for spam email detection. Shifted-1D-LBP, which can be described as an ordered set of binary comparisons of the center value with its neighboring values, is a content-based approach to spam detection with low-level information. To validate the performance of the proposed approach, three benchmark corpora, Spamassasian, Ling-Spam, and TREC email corpuses, were used. The average classification accuracies of the proposed approach were 92.34%, 92.57%, and 95.15%, respectively. Analysis and promising experimental results indicated that the proposed approach was a very competitive feature extraction method in spam email filtering.Öğe Forecasting financial indicators by generalized behavioral learning method(Springer Nature, 2017-08-09) Ertuğrul, Ömer Faruk; Tağluk, Mehmet EminForecasting financial indicators (indexes/prices) is a complex and a quite difficult issue because they depend on many factors such as political events, financial ratios, and economic variables. Also, the psychological facts or decision-making styles of investors or experts are other major reasons for this difficulty. In this study, a generalized behavioral learning method (GBLM) was employed to forecast financial indicators, which are the indexes/prices of 34 different financial indicators (24 stock indexes, 2 forexes, 3 financial futures, and 5 commodities). The achieved results were compared with the reported results in the literature and the obtained results by artificial neural network, which is widely used and suggested for forecasting financial indicators. These results showed that GBLM can be successfully employed in short-term forecasting financial indicators by detecting hidden market behavior (pattern) from their previous values. Also, the results showed that GBLM has the ability to track the fluctuation and the main trend.Öğe Gender classification from facial images using gray relational analysis with novel local binary pattern descriptors(Springer Nature, 2016-11-18) Kaya, Yılmaz; Ertuğrul, Ömer FarukGender classification (GC) is one of the major tasks in human identification that increase its accuracy. Local binary pattern (LBP) is a texture method that employed successfully. But LBP suffers a major problem; it cannot capture spatial relationships among local textures. Therefore, in order to increase the accuracy of GC, two LBP descriptors, which are based on (1) spatial relations between neighbors with a distance parameter, and (2) spatial relations between a reference pixel and its neighbor on the same orientation, were employed to extract features from facial images. Additionally, gray relational analysis (GRA) was carried out to identify gender through extracted features. Experiments on the FEI database illustrated the effectiveness of the proposed approaches. Achieved accuracies are 97.14, 93.33, and 92.50% by applying GRA with the nLBPd, dLBPα, and traditional LBP features, respectively. Experimental results indicated that the proposed approaches were very competitive feature extraction methods in GC. Present work also showed that the nLBPd, dLBPα methods were obtained more acceptable results than traditional LBP.Öğe A novel version of k nearest neighbor: Dependent nearest neighbor(Elsevier, 2017-06) Ertuğrul, Ömer Faruk; Tağluk, Mehmet Emink nearest neighbor (kNN) is one of the basic processes behind various machine learning methods In kNN, the relation of a query to a neighboring sample is basically measured by a similarity metric, such as Euclidean distance. This process starts with mapping the training dataset onto a one-dimensional distance space based on the calculated similarities, and then labeling the query in accordance with the most dominant or mean of the labels of the k nearest neighbors, in classification or regression issues, respectively. The number of nearest neighbors (k) is chosen according to the desired limit of success. Nonetheless, two distinct samples may have equal distances to query but, with different angles in the feature space. The similarity of the query to these two samples needs to be weighted in accordance with the angle going between the query and each of the samples to differentiate between the two distances in reference to angular information. This opinion can be analyzed in the context of dependency and can be utilized to increase the precision of classifier. With this point of view, instead of kNN, the query is labeled according to its nearest dependent neighbors that are determined by a joint function, which is built on the similarity and the dependency. This method, therefore, may be called dependent NN (d-NN). To demonstrate d-NN, it is applied to synthetic datasets, which have different statistical distributions, and 4 benchmark datasets, which are Pima Indian, Hepatitis, approximate Sinc and CASP datasets. Results showed the superiority of d-NN in terms of accuracy and computation cost as compared to other employed popular machine learning methods.Öğe Forecasting electricity load by a novel recurrent extreme learning machines approach(Elsevier, 2016-06) Ertuğrul, Ömer FarukGrowth in electricity demand also gives a rise to the necessity of cheaper and safer electric supply and forecasting electricity load plays a key role in this goal. In this study recurrent extreme learning machine (RELM) was proposed as a novel approach to forecast electricity load more accurately. In RELM, extreme learning machine (ELM), which is a training method for single hidden layer feed forward neural network, was adapted to train a single hidden layer Jordan recurrent neural network. Electricity Load Diagrams 2011-2014 dataset was employed to evaluate and validate the proposed approach. Obtained results were compared with traditional ELM, linear regression, generalized regression neural network and some other popular machine learning methods. Achieved root mean square errors (RMSE) by RELM were nearly twice less than obtained results by other employed machine learning methods. The results showed that the recurrent type ANNs had extraordinary success in forecasting dynamic systems and also time-ordered datasets with comparison to feed forward ANNs. Also, used time in the training stage is similar to ELM and they are extremely fast than the others. This study showed that the proposed approach can be applied to forecast electricity load and RELM has high potential to be utilized in modeling dynamic systems effectively.Öğe Doküman dili tanıma için yeni bir öznitelik çıkarım yaklaşımı: İkili desenler(Gazi Üniversitesi, 2016-12-14) Kaya, Yılmaz; Ertuğrul, Ömer FarukDoğal dil işlemenin önemli alt konularından biri olan dil tanıma (DT), bir dokümanın içeriğine göre yazıldığı dili belirleme işlemidir. Bu çalışmada, karakterlerin UTF-8 değerlerini birbirleri ile karşılaştırmalar sonucu elde edilen ikili desenler kullanarak yeni bir dil tanıma yaklaşımı, bir boyutlu yerel ikili örüntüler (1B-YİÖ) önerilmiştir. Önerilen yöntem farklı sayıda dillerden oluşan metinler içeren dört veri kümesi ile test edilmiştir. 1B-YİÖ ile dokümanlardan elde edilen öznitelikler kullanılarak farklı makine öğrenmesi yöntemleri ile sınıflandırma işlemi gerçekleştirilmiştir. Dört veri kümesi için sınıflandırma başarıları sırası ile %86.20, %92.75, %100 ve %89.77 olarak gözlenmiştir. Elde edilen sonuçlara göre önerilen öznitelik çıkarım yönteminin dil tanıma için önemli örüntüler sağladığı görülmüştür.Öğe A novel approach for SEMG signal classification with adaptive local binary pattern(Springer Nature, 2015-12-31) Ertuğrul, Ömer Faruk; Kaya, Yılmaz; Tekin, RamazanFeature extraction plays a major role in the pattern recognition process, and this paper presents a novel feature extraction approach, adaptive local binary pattern (aLBP). aLBP is built on the local binary pattern (LBP), which is an image processing method, and one-dimensional local binary pattern (1D-LBP). In LBP, each pixel is compared with its neighbors. Similarly, in 1D-LBP, each data in the raw is judged against its neighbors. 1D-LBP extracts feature based on local changes in the signal. Therefore, it has high a potential to be employed in medical purposes. Since, each action or abnormality, which is recorded in SEMG signals, has its own pattern, and via the 1D-LBP these (hidden) patterns may be detected. But, the positions of the neighbors in 1D-LBP are constant depending on the position of the data in the raw. Also, both LBP and 1D-LBP are very sensitive to noise. Therefore, its capacity in detecting hidden patterns is limited. To overcome these drawbacks, aLBP was proposed. In aLBP, the positions of the neighbors and their values can be assigned adaptively via the down-sampling and the smoothing coefficients. Therefore, the potential to detect (hidden) patterns, which may express an illness or an action, is really increased. To validate the proposed feature extraction approach, two different datasets were employed. Achieved accuracies by the proposed approach were higher than obtained results by employed popular feature extraction approaches and the reported results in the literature. Obtained accuracy results were brought out that the proposed method can be employed to investigate SEMG signals. In summary, this work attempts to develop an adaptive feature extraction scheme that can be utilized for extracting features from local changes in different categories of time-varying signals.Öğe A novel type of activation function in artificial neural networks: Trained activation function(Elsevier, 2018-03) Ertuğrul, Ömer FarukDetermining optimal activation function in artificial neural networks is an important issue because it is directly linked with obtained success rates. But, unfortunately, there is not any way to determine them analytically, optimal activation function is generally determined by trials or tuning. This paper addresses, a simpler and a more effective approach to determine optimal activation function. In this approach, which can be called as trained activation function, an activation function was trained for each particular neuron by linear regression. This training process was done based on the training dataset, which consists the sums of inputs of each neuron in the hidden layer and desired outputs. By this way, a different activation function was generated for each neuron in the hidden layer. This approach was employed in random weight artificial neural network (RWN) and validated by 50 benchmark datasets. Achieved success rates by RWN that used trained activation functions were higher than obtained success rates by RWN that used traditional activation functions. Obtained results showed that proposed approach is a successful, simple and an effective way to determine optimal activation function instead of trials or tuning in both randomized single and multilayer ANNs.Öğe Extreme learning machine model for water network management(Springer Nature, 2017-04-22) Sattar, Ahmed M.A.; Ertuğrul, Ömer Faruk; Gharabaghi, Bahram; McBean, Edward Arthur; Cao, JiuwenA novel failure rate prediction model is developed by the extreme learning machine (ELM) to provide key information needed for optimum ongoing maintenance/rehabilitation of a water network, meaning the estimated times for the next failures of individual pipes within the network. The developed ELM model is trained using more than 9500 instances of pipe failure in the Greater Toronto Area, Canada from 1920 to 2005 with pipe attributes as inputs, including pipe length, diameter, material, and previously recorded failures. The models show recent, extensive usage of pipe coating with cement mortar and cathodic protection has significantly increased their lifespan. The predictive model includes the pipe protection method as pipe attributes and can reflect in its predictions, the effect of different pipe protection methods on the expected time to the next pipe failure. The developed ELM has a superior prediction accuracy relative to other available machine learning algorithms such as feed-forward artificial neural network that is trained by backpropagation, support vector regression, and non-linear regression. The utility of the models provides useful inputs when planning and budgeting for watermain inspection, maintenance, and rehabilitation.Öğe Two novel versions of randomized feed forward artificial neural networks: Stochastic and pruned stochastic(Springer Nature, 2017-11-13) Ertuğrul, Ömer FarukAlthough high accuracies were achieved by artificial neural network (ANN), determining the optimal number of neurons in the hidden layer and the activation function is still an open issue. In this paper, the applicability of assigning the number of neurons in the hidden layer and the activation function randomly was investigated. Based on the findings, two novel versions of randomized ANNs, which are stochastic, and pruned stochastic, were proposed to achieve a higher accuracy without any time-consuming optimization stage. The proposed approaches were evaluated and validated by the basic versions of the popular randomized ANNs [1] are the random weight neural network [2], the random vector functional links [3] and the extreme learning machine [4] methods. In the stochastic version of randomized ANNs, not only the weights and biases of the neurons in the hidden layer but also the number of neurons in the hidden layer and each activation function were assigned randomly. In pruned stochastic version of these methods, the winner networks were pruned according to a novel strategy in order to produce a faster response. Proposed approaches were validated via 60 datasets (30 classification and 30 regression datasets). Obtained accuracies and time usages showed that both versions of randomized ANNs can be employed for classification and regression.
