Makale Koleksiyonu

Bu koleksiyon için kalıcı URI

https://hdl.handle.net/11491/3174

Listeleniyor 1 - 15 / 15

Accurate detection of coronavirus cases using deep learning with attention mechanism and genetic algorithm
(SPRINGER, 2024) Kara, Ahmet
The novel coronavirus disease has caused severe threats to the daily life and health of people all over the world. Hence, early detection and timely treatment of this disease are signifcant to prevent the coronavirus’s spread and ensure more efective patient care. This work adopted an integrated framework comprising deep learning and attention mechanism to provide a more efective and reliable diagnosis. This framework consists of two convolution neural network (CNN), a bidirectional LSTM, two fully-connected layers (FCL), and an attention mechanism. The main aim of the proposed framework is to reveal a promising approach based on deep learning for early and timely detection of coronavirus disease. For greater accuracy, the framework’s hyperparameters are tuned by means of a genetic algorithm. The efectiveness of the proposed framework has been examined utilizing a public dataset including 18 diferent blood fndings from Albert Einstein Israelita Hospital in Sao Paulo, Brazil. Additionally, within the experimental studies, the proposed framework is subjected to comparison with the state-of-the-art techniques, evaluated across various metrics. Based on the derived consequences, the proposed framework has yielded enhancements in accuracy, recall, precision, and F1-score, registering approximate improvements of 1.27%, 4.07%, 3.20%, and 2.88%, respectively, as measured against the second-best rates.
Shortest Confidence Intervals of Weibull Modulus for Small Samples in Materials Reliability Analysis
(GAZI UNIV, 2023) Yalçınkaya, Meryem; Birgören, Burak
The Weibull distribution has been widely used to model strength properties of brittle materials. Estimation of confidence intervals for Weibull shape parameter has been an important concern, since small sample sizes in materials science experiments bring about large intervals. Many methods have been proposed in the literature for constructing shorter intervals; the methods of maximum likelihood, least square, and Menon are among the most extensively studied methods. However, they all use an equal-tails approach. The pivotal quantities used for constructing confidence intervals have right-skewed and unimodal distributions, thus, they clearly do not produce the shortest intervals for a given confidence level in equal tail form. This study constructs the shortest confidence intervals for the three aforementioned methods and compares their performances by their equal-tails counterparts. To this end, a comprehensive simulation study has been conducted for the shape parameter values between 1 to 80 and the sample sizes between 3 to 20. The comparison criterion is chosen as the expected interval length. The results show that the shortest confidence intervals in each of three methods have yielded considerably narrower intervals. Further, the unknown parameter values are more centered in these intervals.
Deep Learning Approach to Technician Routing and Scheduling Problem
(EDICIONES UNIV SALAMANCA, 2022) Pekel, Engin
This paper proposes a hybrid algorithm including the Adam algorithm and body change operator (BCO). Feasible solutions to technician routing and scheduling problems (TRSP) are investigated by performing deep learning based on the Adam algorithm and the hybridization of Adam-BCO. TRSP is a problem where all tasks are routed, and technicians are scheduled. In the deep learning method based on the Adam algorithm and Adam-BCO algorithm, the weights of the network are updated, and these weights are evaluated as Greedy approach, and routing and scheduling are performed. The performance of the Adam-BCO algorithm is experimentally compared with the Adam and BCO algorithm by solving the TRSP on the instances developed from the literature. The numerical results evidence that Adam-BCO offers faster and better solutions considering Adam and BCO algorithm. The average solution time increases from 0.14 minutes to 4.03 minutes, but in return, Gap decreases from 9.99% to 5.71%. The hybridization of both algorithms through deep learning provides an effective and feasible solution, as evidenced by the results.
Uzun-Kısa Süreli Bellek Ağı Kullanarak Global Güneş Işınımı Zaman Serileri Tahmini
(2019) Kara, Ahmet
Global güneş ışınımı tahmini, güneş enerjisi sistemlerinin etkin yönetimi ve işletilmesinin yanı sıra gelecekteki enerji üretimi hakkında güvenilir bilgi sağlamak için giderek daha fazla önem kazanmaktadır. Bu çalışmada, günlük güneş ışınım tahmin problemini etkin bir model oluşturmak için Uzun Kısa Süreli Bellek (Long Short-Term Memory - LSTM) ağı önerilmiştir. Önerilen yöntemin etkinliği Karar Ağaçları Regresyon, Rastgele Orman Regresyon, Gradyan Güçlendirme ve K-En Yakın Komşu gibi en etkili makine öğrenme algoritmalar ile karşılaştırılmıştır. LSTM modelinin yaklaşımının etkinliğini doğrulamak için Çorum - Türkiye’de Temmuz-1983 ve Aralık-2018 tarihleri arasında global güneş ışınımı sıralı zaman serileri verileri kullanılmıştır. Simülasyon sonuçları, LSTM yönteminin diğer makine öğrenme modellerinden daha iyi performansa sahip olduğunu göstermektedir.
A data-driven approach based on deep neural networks for lithium-ion battery prognostics
(Springer London Ltd, 2021) Kara, Ahmet
Remaining useful life estimation is gaining attention in many real-world applications to alleviate maintenance expenses and increase system reliability and efficiency. Deep learning approaches have recently provided a significant improvement in the estimation of remaining useful life (RUL) and degradation progression concerning machinery prognostics. This research presents a new data-driven approach for RUL estimation using a hybrid deep neural network that combines CNN, LSTM, and classical neural networks. The presented CNN-LSTM neural network aims to extract the spatio-temporal relations in multivariate time series data and capture nonlinear characteristics to achieve better RUL prediction accuracy. To improve the proposed model's performance, PSO is handled to simultaneously optimize the hyperparameters of the network consisting of the number of epochs, the number of convolutional and LSTM layers, the size of units (or filters) in each convolutional, and LSTM layers. Besides, the proposed model in this paper, called the CNN-LSTM-PSO, realizes the multi-step-ahead prediction. In the experimental studies, the popular lithium-ion battery dataset presented by NASA is selected to verify the CNN-LSTM-PSO approach. The experimental consequences revealed that the presented CNN-LSTM-PSO model gives better results than other state-of-the-art machine learning techniques and deep learning approaches considering various performance criteria.
Multi-step influenza outbreak forecasting using deep LSTM network and genetic algorithm
(Pergamon-Elsevier Science Ltd, 2021) Kara, Ahmet
Influenza epidemic is a serious public health problem that has attracted worldwide attention due to cause cases of severe illness, an enormous economic burden, and even deaths worldwide each year. Forecasting influenza outbreak in advance has great significance on influenza-like illness (ILI) prevention and healthcare management. Existing research approaches based on traditional statistical and machine learning have failed to select superior features that detect sophisticated and non-linear characteristics of influenza epidemic sequential data. In this paper, it is introduced a hybrid method that combines long short-term memory (LSTM) neural network and genetic algorithm (GA) for multi-step influenza outbreak forecasting problems. LSTM model is employed to overcome the complexity and nonlinearity issues in an influenza prediction. In order to enhance the efficiency and performance of the neural network, the genetic algorithm is used to obtain the epoch size of the network, the number of LSTM layers, the size of units in each LSTM layer, and the time window size simultaneously. For comparison purposes, it is chosen the weekly data of influenza-like illness (ILI), also known as influenza or other similar illness showing flu-like symptoms, in the USA collected by the Centers for Disease Control and Prevention (CDC). The experimental results demonstrated that the presented hybrid model outperforms other highly developed machine learning approaches, a statistical model, and a fully-connected neural network considering different performance metrics during peak periods.
Karma modelli tip-2 montaj hattı dengeleme problemi için bir kısıt programlama modeli
(Pamukkale Üniversitesi, 2016) Alağaş, Hacı Mehmet; Pınarbaşı, Mehmet; Yüzükırmızı, Mustafa; Toklu, Bilal
Bu çalışmada karma modelli montaj hattı dengeleme problemleri için yeni bir kısıt programlama modeli sunulmuştur. Önerilen model verilen bir istasyon sayısı ile çevrim zamanını en küçüklemektedir. Önerilen model literatürdeki örnek problemler ile test edilmiştir ve modelin performansı karma modelli montaj hattı dengeleme problemlerinin matematiksel modeli ile karşılaştırmalı olarak değerlendirilmiştir. Performans kriterleri olarak ulaşılan en iyi çözüm değeri ve CPU süresi kullanılmıştır. Deneysel sonuçlar önerilen kısıt programlama modelinin problemin çözümünde iyi performans gösteren bir alternatif modelleme tekniği olduğunu göstermiştir.
Analysis of bowl effects on assembly line using queueing networks and constraint programming procedure
(Computers and Industrial Engineering, 2014) Pınarbaşı, Mehmet; Alağaş, Hacı Mehmet; Yüzükırmızı, Mustafa; Toklu, Bilal
In this study, a new solution procedure based on queueing networks and constraint programming is proposed to model and solve the Assembly Line Balancing Problem (ALBP). Variation of the task and the station times, and precedence relation effects are considered to evaluate the line performance. Station utilization, total average number of jobs and smoothness index are used as performance measures. Bowl effect, inverted bowl effect and variability imbalance which are seen in balanced lines are examined by using proposed procedure. Also effects of the variability on the line performance are reviewed. Literature data sets are utilized to assess the effectiveness of the procedure.
A new simulated annealing approach for travelling salesman problem
(2013) Bayram, Hüsamettin; Şahin, Ramazan
The aim of this study is to improve searching capability of simulated annealing (SA) heuristic through integration of two new neighborhood mechanisms. Due to its ease of formulation, difficulty to solve and various real life applications several Travelling Salesman Problems (TSP) were selected from the literature for the testing of the proposed methods. The proposed methods were also compared to conventional SA with swap neighborhood. The results have shown that the proposed techniques are more effective than conventional SA, both in terms of solution quality and time.
Design process model for flexible manufacturing systems and application of TAI
(2011) Pınarbaşı, Mehmet; Yüzükırmızı, Mustafa
In this paper, a systematic design process model is proposed to model Flexible Manufacturing Systems (FMS). Generally, FMS systems design is complicated and cost very high as installation phase. Therefore, technical analysis of system design should be exact and accurate even for the installation phase. At this stage, the proposed model is Integration Definition for Function Modeling (IDEF) diagram which is a structured design technique. Proposed model is exemplified with a practical example in TAI which is operating in the aerospace industry. The performance of the proposed model is analyzed using simulation and queueing network technique. In the performance analysis, test parameters are determined as service rates, number of servers, routing information and number of transporters and also performance parameters are determined cycle time and machine utilization.
A comprehensive mathematical model for dynamic cellular manufacturing system design and Linear Programming embedded hybrid solution techniques
(Elsevier Ltd, 2016) Bayram, Hüsamettin; Şahin, Ramazan
Considering the ever changing market conditions, it is essential to design responsive and flexible manufacturing systems. This study addresses the multi-period Dynamic Cellular Manufacturing System (DCMS) design problem and introduces a new mathematical model. The objective function of the mathematical model considers inter-cell and intra-cell material handling, machine purchasing, layout reconfiguration, variable and constant machine costs. Machine duplication, machine capacities, operation sequences, alternative processing routes of the products, varying demands of products and lot splitting are among the most important issues addressed by the mathematical model. It makes decisions on many system related issues, including cell formation, inter- and intra-cell layout, product routing and product flow between machines. Due to the complexity of the problem, we suggest two heuristic solution approaches that combine Simulated Annealing (SA) with Linear Programming and Genetic Algorithm (GA) with Linear Programming. The developed approaches were tested using a data set from the literature. In addition, randomly generated test problems were also used to investigate the performance of the hybrid heuristic approaches. A problem specific lower bound mathematical model was also proposed to observe the solution quality of the developed approaches. The suggested approaches outperformed the previous study in terms of both computational time and the solution quality by reducing the overall system cost. © 2015 Elsevier Ltd. All rights reserved.
Solving capacitated location routing problem by variable neighborhood descent and GA–Artificial neural network hybrid method
(Faculty of Transport and Traffic Engineering, 2018) Pekel, Engin; Soner Kara, Selin
This paper aims to find the optimal depot locations and vehicle routings for spare parts of an automotive company considering future demands. The capacitated location-routing problem (CLRP), which has been practiced by various methods, is performed to find the optimal depot locations and routings by additionally using the artificial neural network (ANN). A novel multi-stage approach, which is performed to lower transportation cost, is carried out in CLRP. Initially, important factors for customer demand are tested with an univariate analysis and used as inputs in the prediction step. Then, genetic algorithm (GA) and ANN are hybridized and applied to provide future demands. The location of depots and the routings of the vehicles are determined by using the variable neighborhood descent (VND) algorithm. Five neighborhood structures, which are either routing or location type, are implemented in both shaking and local search steps. GA-ANN and VND are applied in the related steps successfully. Thanks to the performed VND algorithm, the company lowers its transportation cost by 2.35% for the current year, and has the opportunity to determine optimal depot locations and vehicle routings by evaluating the best and the worst cases of demand quantity for ten years ahead. © 2018, Faculty of Transport and Traffic Engineering. All rights reserved.
Integrated definition modeling and Taguchi analysis of flexible manufacturing systems: Aircraft industry application
(2013) Pınarbaşı, Mehmet; Sel, Çağrı; Alağaş, Hacı Mehmet; Yüzükırmızı, Mustafa
In this paper, flexible manufacturing systems (FMS) are studied. Firstly, an FMS design approach is proposed using integrated definition for function methodology. A systematic layout design and performance evaluation scheme is presented and detailed using this modeling framework. Then, the proposed approach is carried out with a case study from an aircraft industry to convert an existing traditional production system to FMS. To improve the system performance, a simulation-based method with Taguchi approach consisting of multiproducts is utilized. The objective is to find the machine and the product mix that achieves the maximum utilization while minimizing the cycle time. FMS system performance has been greatly improved by determining the most advantageous level of system components. It has also shown that FMS is a practicable production system in aircraft industry. © 2013 The Author(s).
MLSeq: Machine learning interface for RNA-sequencing data
(Elsevier Ireland Ltd, 2019) Göksuluk, Dinçer; Zararsız, Gökmen; Korkmaz, Selçuk; Eldem, Vahap; Ertürk Zararsız, Gözde; Özçetin, Erdener; Öztürk, Ahmet; Karaağaoğlu, Ahmet Ergun
Background and Objective: In the last decade, RNA-sequencing technology has become method-of-choice and prefered to microarray technology for gene expression based classification and differential expression analysis since it produces less noisy data. Although there are many algorithms proposed for microarray data, the number of available algorithms and programs are limited for classification of RNA-sequencing data. For this reason, we developed MLSeq, to bring not only frequently used classification algorithms but also novel approaches together and make them available to be used for classification of RNA sequencing data. This package is developed using R language environment and distributed through BIOCONDUCTOR network. Methods: Classification of RNA-sequencing data is not straightforward since raw data should be preprocessed before downstream analysis. With MLSeq package, researchers can easily preprocess (normalization, filtering, transformation etc.) and classify raw RNA-sequencing data using two strategies: (i) to perform algorithms which are directly proposed for RNA-sequencing data structure or (ii) to transform RNA-sequencing data in order to bring it distributionally closer to microarray data structure, and perform algorithms which are developed for microarray data. Moreover, we proposed novel algorithms such as voom (an acronym for variance modelling at observational level) based nearest shrunken centroids (voomNSC), diagonal linear discriminant analysis (voomDLDA), etc. through MLSeq. Materials: Three real RNA-sequencing datasets (i.e cervical cancer, lung cancer and aging datasets) were used to evalute model performances. Poisson linear discriminant analysis (PLDA) and negative binomial linear discriminant analysis (NBLDA) were selected as algorithms based on dicrete distributions, and voomNSC, nearest shrunken centroids (NSC) and support vector machines (SVM) were selected as algorithms based on continuous distributions for model comparisons. Each algorithm is compared using classification accuracies and sparsities on an independent test set. Results: The algorithms which are based on discrete distributions performed better in cervical cancer and aging data with accuracies above 0.92. In lung cancer data, the most of algorithms performed similar with accuracies of 0.88 except that SVM achieved 0.94 of accuracy. Our voomNSC algorithm was the most sparse algorithm, and able to select 2.2% and 6.6% of all features for cervical cancer and lung cancer datasets respectively. However, in aging data, sparse classifiers were not able to select an optimal subset of all features. Conclusion: MLSeq is comprehensive and easy-to-use interface for classification of gene expression data. It allows researchers perform both preprocessing and classification tasks through single platform. With this property, MLSeq can be considered as a pipeline for the classification of RNA-sequencing data. © 2019 Elsevier B.V.
Variability modelling and balancing of stochastic assembly lines
(Taylor and Francis Ltd., 2016) Pınarbaşı, Mehmet; Yüzükırmızı, Mustafa; Toklu, Bilal
In a production flow line with stochastic environment, variability affects the system performance. These stochastic nature of real-world processes have been classified in three types: arrival, service and departure process variability. So far, only service process – or task time – variation has been considered in assembly line (AL) balancing studies. In this study, both service and flow process variations are modelled along with AL balancing problem. The best task assignment to stations is sought to achieve the maximal production. A novel approach which consists of queueing networks and constraint programming (CP) has been developed. Initially, the theoretical base for the usage of queueing models in the evaluation of AL performance has been established. In this context, a diffusion approximation is utilised to evaluate the performance of the line and to model the variability relations between the work stations. Subsequently, CP approach is employed to obtain the optimal task assignments to the stations. To assess the effectiveness of the proposed procedure, the results are compared to simulation. Results show that, the procedure is an effective solution method to measure the performance of stochastic ALs and achieve the optimal balance. © 2016 Informa UK Limited, trading as Taylor & Francis Group.

Güncel Gönderiler