Sürdürülebilir Kalkınma Amaçları
Yazarlar
Yayınlar
- MakaleAçık Erişim2026Scopus SJRQ2Using Decision Tree and K-Means to Improve ANFIS for Predicting Missing Flow Data in Çoruh Basin
Acar, Ramazan
Complete and continuous flow data provides contributions to the water resources projects appropriate design. However, both financial and technical factors prevent the regular collection of data and such a situation leads to missing data problems. Input variables are from five stations in the Çoruh Basin in the northeast province of Türkiye, and output variables are from one station. In the study, 75% (171 months of data) of the 228 months flow data between the years 1993–2011 are used in the training phase and 25% (57 months of data) in the testing phase. In the ANFIS models, Decision Tree (DT) and K-means (KM) methods are used in selecting the input variables and determining the number of membership functions. In addition, artificial neural network (ANN), multiple linear regression (MLR), and ANFIS models with randomly determined cluster numbers are also used to compare the model performance of these methods. The results show that using DT and KM methods, the ANFIS models generate more reliable results than other models. During the study, the regression coefficient (R2), weighted mean square error (WMSE), and Wilcoxon (Z) values are taken into account for reliability measure. The DT–K-means–ANFIS model achieved the highest accuracy with R2 = 0.98 and WMSE = 5.89 during the testing phase, demonstrating superior performance to the ANN and MLR models. It is shown that the most successful models can be determined in a shorter time by using DT and KM methods before generating ANFIS models.
AtıfScopus1WoS1OpenAlex1
İncele - MakaleAçık Erişim2024Scopus SJRQ3Performance of artificial neural networks and traditional methods in determining selected growth parameters of Alburnus sellal Heckel, 1843
Ozcan, Ebru Ifakat
In this study, predictions were made on the growth performance of Alburnus sellal Heckel, 1843 from the Munzur River using back propagation artificial neural networks and ANN algorithms. Statistical growth models used in fish biology and results obtained from artificial neural networks were compared. A total of 239 samples were used in this comparison. It was determined that the population is distributed across age groups 0–VII. The relationship between length and weight was calculated as W = 0.0046L3.198 for all individuals. The von Bertalanffy growth parameters were calculated for all individuals: Lt = 21.93 [1 – e–0.158 (t + 2.11)]; Wt = 102.29 [1 – e–0.158 (t + 2.11)]3.198. The growth performance index (Ф’) value was 1.880 for all individuals. The condition factor varied between 0.479 and 1.115 for females and between 0.533 and 1.076 for males. The Mean Absolute Percent Error (MAPE) statistic was used, which is a widely used method to measure the accuracy of the predictions made. It was determined that ANNs MAPE (%) values were better than MAPE values calculated for the length–weight relationship and von Bertalanffy growth function models for A. sellal. This study shows that ANNs can be used as an alternative useful method for predicting population parameters. ANN models are therefore an effective tool to describe fish growth parameters. They have been found to be a useful predictive tool. The developed models can be used to predict future sustainable fish management.
AtıfScopus2WoS3OpenAlex1İncele - MakaleYalnız künye2020Scopus SJRQ1Emergency department network under disaster conditions: The case of possible major Istanbul earthquake
Gul, Muhammet; Fuat Guneri, Ali; Gunal, Murat M.
Emergency departments (EDs) provide health care services to people in need of urgent care. Their role is remarkable when extraordinary events that affect the public, such as earthquakes, occur. In this paper, we present a hybrid framework to evaluate earthquake preparedness of EDs in cities. Our hybrid framework uses artificial neural networks (ANNs) to estimate number of casualties and discrete event simulation (DES) to analyse the effect of surge in patient demand in EDs, after an earthquake happens. At the core of our framework, Earthquake Time Emergency Department Network Simulation Model (ET-EDNETSIM) resides which can simulate patient movements in a network of multiple and coordinated EDs. With the design of simulation experiments, different resource levels and sharing rules between EDs can be evaluated. We demonstrated our framework in a network of five EDs located in a region of which is estimated to have the highest injury rate after an earthquake in Istanbul, Turkey. Results of our study contributed to the planning for expected earthquake in Istanbul. Simulating a network of EDs extends the individual ED studies in the literature and furthermore, our hybrid framework can help increase earthquake preparedness in cities around the world. On the methodological side, the use of ANN, which is a member of machine learning (ML) algorithms family, in our hybrid framework also shows the close links between ML and DES.
AtıfScopus30WoS31OpenAlex40İncele - MakaleYalnız künye2019Scopus SJRQ4Evaluation of a new computer method (ANNs) and traditional methods (LWRS and VBGF) in the calculation of some growth parameters of two cyprinid species
Ozcan, Ebru Ifakat; Serdar, Osman
Artificial Neural Networks (ANNs) are a very reliable computer program used in many areas such as growth parameters. The aim of this study to compare the growth estimates of tigris scraper (Capoeta umbla (Heckel, 1843)) and kura barbel (Barbus lacerta (Heckel, 1843) captured from Karasu River (East Anatolia, Turkey) between 2014 and 2016. The age groups were 0-9 for C. umbla and 0-8 for B. lacerta. The estimated parameters of von Bertalanffy growth equations were as L' =51.96 cm, W' 804.16 g, k=0.092 year-1, t0=-1.763 year for C. umbla and L' 32.23 FP, W' 435.87 J, k 0.127\HDU-1, t0-2.489 for B. lacerta. The growth performance index (Y) value was comput-ed as 2.395 for C. umbla and 2.120 for B. lacerta. The length and weight relationships for two fish species were as follows: W=0.0092L3.05 (R²=0.98) and W=0.0095L3.07 (R2=0.98). The growth of these two cyprinid species were isometric growth. A ANNs model is a interdepented system containing many neurons. There are three different layers in a neural network (input, hidden and output layers). Data enters (length/weight, age, sex) at the inputs and passes along the network, by layer, till it reach-es at the outputs (weight/length). MATLAB Ver R2016a (Neural Network Toolbox) was used for ANNs estimatings. MAPE (%) equation was used to compare ANNs and other methods. According to the comparison of the results obtained with MAPE (%), ANNs provide better results than the LWRs and VBGF. The results of this study could give useful insight that ANNs can be evaluated as an alternative for growth prediction.
AtıfScopus9İncele - MakaleYalnız künye2018Scopus SJRQ3ARTIFICIAL NEURAL NETWORKS AS NEW ALTERNATIVE METHOD TO ESTIMATING SOME POPULATION PARAMETERS OF TIGRIS LOACH (OXYNOEMACHEILUS TIGRIS (HECKEL, 1843)) IN THE KARASU RIVER, TURKEY
Ozcan, Ebru Ifakat; Serdar, Osman
In this study, a new method (Artificial Neural Networks) to estimating some population parameters of Tigris loach (Oxynoemacheilus tigris (Heckel, 1843)) collected between 2014 and 2015 in 14 different stations from Karasu River (East Anatolia, Turkey). The total length and weight of the species ranged from 62 to105 mm and from 1.7 to 10.9 g for females and 65 to 100 mm and from 2.4 to 9.8 g for males, respectively. The new maximum length for O. tigris was found to be 105 mm in this study. The length-weight relation of O. tigris was carried out with artificial neural networks (ANNs) and length-weight relationships (LWRs). The results obtained were compared. Length-weight relationships of O. tigris were found as W=0.0049*L-3.29 , R-2 =0.95 for all individuals. According to these values, the growth type of this species was positive allometric (b>3) for all individuals. Regression analysis showed that fish length has high significant correlation with weight (R=0.97, R-2 =0.95, F-1,F-185=31930.3, P<0.001). According to the comparison of the results obtained with MAPE (%), ANNs provide better results than the LWRs. The results of this study could give useful insight that ANNs can be evaluated as an alternative for length-weight relation estimation. The condition factor values (min-max) were observed as 0.680-1.239 for females and 0.699-1.074 for males. No studies have been done before on this species population parameters.
AtıfWoS15İncele - MakaleAçık Erişim2018Scopus SJRQ1Artificial neural networks and adaptive neuro-fuzzy inference systems approaches to forecast the meteorological data for HVAC: The case of cities for Turkey
Işık, Erdem; Inallı, Mustafa
Limited energy resources and increasing need for energy due to population growth seem to lead researchers to focus on these issues. Forecast of meteorological data has significant importance in design of thermal systems. In this study, forecasting of meteorological data used in thermal system design was performed for fifty cities to represent the entire Turkey. Data obtained from General Directorate of Meteorology (MGM) were modelled by artificial neural networks and adaptive-network based fuzzy inference systems. Matlab software was used for modeling and forecasting of prospective data with high sensitivity in thermal systems. Surfer and ArcGIS software were used to create humidity, temperature, solar radiation maps for Turkey. Root mean square error (RMSE), mean absolute error and (MAE), coefficient of variation (COV) and the coefficient of determination (R2) were used to validate the result of the proposed approaches. The results were satisfactory with respect to RMSE, MAE, COV and R2 to forecast the meteorological data. Annual solar power potential maps for Turkey were also proposed and compared with MGM results. The results of the proposed approaches are compatible with the result of the MGM. The Turkey policy maker(s) from MGM can easily use these approaches if a software is constructed.
AtıfScopus66WoS52OpenAlex68İncele - BildiriYalnız künye2017An ann and ga approach for demand forecasting and routing for cash management
Gumus, Alev Taskin; Celik, Erkan; Ömerustaoğlu, Furkan; Drobne, S.; Zerovnik, J.; Zadnik Stirn, L.; Kljajic Borstar, M.
Forecasting of cash demand and cash transportation are critical processes for all bank asynchronous transfer modes (ATMs). In this paper, an application is presented to accurate forecasting of cash demand and minimizing the transportation cost. In the first step of the proposed application, artificial neural network is implemented for forecasting the amount of withdrawn and deposited money from an ATM in one day. Hence, a more accurate way is provided to determine the amount of money that will be loaded into the ATM, and opportunity costs are minimized. Then, we applied genetic algorithm for minimizing the transportation cost of cash deposit to the ATMs.
- MakaleAçık Erişim2017Scopus SJRQ3Application of artificial neural networks using bayesian training rule in sales forecasting for furniture industry; [Primjena umjetnih neuronskih mreža uz pomoć Bayesova pravila učenja u predviđanju prodaje za industriju namještaja]
Yucesan, Melih; Gul, Muhammet; Celik, Erkan
Most organizations in manufacturing environments aim to increase their profits and reduce costs against competitive and rapidly changing market conditions. Accuracy of sales forecasting is undoubtedly a successful way to reach the aforementioned goals. At the same time, this enables executives to improve customer satisfaction, reduce lost sales and plan production efficiently. As a growing industry in Turkey, furniture manufacturing has an increased product demand in relation to the recent growth in construction and related industries, increase in urban population and increase in person-level income. Therefore, accurate sales forecasting systems in this industry are more focused on the special and calendar factors, such as consumer confidence index, producer price index, time of the year and number of vacation days. In this paper, an artificial neural network (ANN) based forecasting model is proposed by using MATLAB for processing total monthly sales data of a corporate furniture manufacturer located in the Black Sea region of Turkey. The method is a component of ANN, namely Bayesian regularization. The proposed model is applied to monthly sales figures of a corporate furniture manufacturing company. In conclusion, the results of performance measures show that using the ANN model based on Bayesian rules training is an applicable choice for forecasting of monthly sales of the observed furniture factory.
AtıfScopus14WoS11İncele