Sürdürülebilir Kalkınma Amaçları
Yazarlar
- Aksarayli, Mehmet(1)
- Altuntas, Serkan(1)
- Arslan, Hakan(1)
- Arslan, Rukiye Uzun(1)
- Boz, Ediz(1)
- Elbir, Tugba(1)
- Engin, Mustafa Alptekin(1)
- Kamiloglu, Aybike(1)
- Kamiloglu, Hakan Alper(1)
- Karpuz, Ceyhun(1)
- Maras, Sinan(1)
- Tekin, Duygu(1)
- Tekin, Ilker(1)
- Togan, Vedat(1)
- Topçu, Kubra Cinar(1)
- Turan, Huseyin(1)
- Türk, Ersin(1)
- Yapici, Irem Senyer(1)
- Çakir, Mehmet(1)
- Öztekin, Ertekin(1)
Yayınlar
- MakaleAçık Erişim2026Improvement of machine learning-based diabetes diagnosis via resampling techniques
Yapici, Irem Senyer; Arslan, Rukiye Uzun; Engin, Mustafa Alptekin
The objective of this study is to enhance the accuracy of diabetes diagnosis through the utilisation of machine learning techniques and resampling methods. The imbalanced nature of diabetes datasets presents a significant challenge for traditional classification algorithms, which often struggle to accurately predict results. In order to enhance the efficacy of the model, a comparative analysis was conducted to assess the performance of a range of over-sampling and under-sampling techniques, including SMOTE, ADASYN, Borderline SMOTE, SVM SMOTE, Random Under Sampler, Near Miss, One Sided Selection, Neighbourhood Cleaning Rule, Edited Nearest Neighbours, Instance Hardness Threshold, AllKNN and Tomek Links. The aforementioned techniques were then applied to the Decision Tree, Random Forest, K-Nearest Neighbours, AdaBoost, Extra Tree Classifier, and machine learning classifiers, and their performance was evaluated using the accuracy, recall, precision, F-Score, and AUC-ROC performance metrics. The SVMSMOTE resampling technique was identified as the most successful method, achieving 99.06% accuracy when used in combination with the decision tree classifier. The findings demonstrate that the incorporation of resampling techniques markedly enhances diagnostic proficiency and yields more dependable forecasts. This research makes a significant contribution to the field of medical informatics, providing a robust framework for diabetes diagnosis and offering valuable insights into the application of machine learning healthcare.
- MakaleAçık Erişim2025Two-and four-way Wilkinson power divider design with slow-wave microstrip structures
Çakir, Mehmet; Karpuz, Ceyhun
In this study, conventional Wilkinson power dividers have been modified with meandered transmission line structures, resulting in two-port power dividers with either two or three serial connections and four-port power dividers with two serial connections. A substrate with a relative permittivity of 10.2 and a height of 1.27 mm, RT/Duroid, was used for the designs, and frequency responses were optimized using an electromagnetic simulation program. To enhance isolation and bandwidth, 100-ohm resistors were placed at each connection point in the power dividers. In the two-port configuration with two serial connections, the average return loss was-14.45 dB, insertion loss was-3.40 dB, and isolation was-13.25 dB over the 0.17-1.36 GHz band. Additionally, in the 2.24-2.97 GHz frequency band, the return loss was-15.45 dB, insertion loss was-3.50 dB, and isolation was-15.93 dB. In the three-port configuration with three serial connections, the average return loss was-14.73 dB, insertion loss was-3.38 dB, and isolation was-12.20 dB over the 0.11-1.05 GHz range;-15.21 dB return loss,-3.54 dB insertion loss, and-15.34 dB isolation over the 1.36-2.10 GHz range; and-17.52 dB return loss,-3.65 dB insertion loss, and-14.03 dB isolation over the 2.59-3.14 GHz range. In the four-port configuration with two serial connections, the average return loss was-20.65 dB, insertion loss was-6.17 dB, and isolation was-16.74 dB in the 0.47-1.17 GHz range. The designed circuits were fabricated, and it was observed that the simulation results were consistent with the measurement results.
AtıfWoS1TR Dizin1
İncele - MakaleAçık Erişim2023Prioritising and selecting the railway routes planned in the Eastern Black Sea Region with the AHP and TOPSIS methods
Boz, Ediz; Türk, Ersin; Togan, Vedat
The construction of the railway, which is needed for the development of the Eastern Black Sea Region, is an issue that has been discussed and unresolved for about 110 years. The geomorphological structure of the region and the developments in the highway played a role in the delay of the railway construction. However, the developing technology in the railway, Turkey's being a part of the Iron Silk Road Project, and the agreements made with international connections have brought the railway construction to the region again. In the Environmental Plan (EP) approved and in force by the Ministry of Environment, Urbanization and Climate Change, four railway routes are planned to integrate the Eastern Black Sea region with the national/international railway network. These routes are; Samsun-Sarp, Trabzon-Erzincan, Of -Erzurum and Tirebolu-Bayburt. However, some of these routes are alternatives to each other and are not feasible and rational since the implementation of all planned routes requires great costs. The aim of this study is to develop a method for prioritizing and selecting the railway routes planned in the EP. Within the scope of this purpose, by using Analytical Hierarchy Process (AHP) and TOPSIS techniques, which are multi-criteria decision making methods, the routes were listed according to 4 criteria and 10 sub-criteria determined with the help of literature and expert team. As a result of the study; In both techniques, the order was the same and it was determined that the Samsun-Sarp railway was the most priority railway route. Trabzon-Erzincan came second, Of-Erzurum came third and Tirebolu-Bayburt came fourth.
AtıfWoS3OpenAlex3TR Dizin2İncele - MakaleAçık Erişim2021Effect of tarragon addition on volatile compound profile and some quality paramaters of sucuk
Kamiloglu, Aybike; Elbir, Tugba; Topçu, Kubra Cinar
The goal of the present study was to investigate the effect of tarragon and ripening period on certain quality parameters and the volatile profile of sucuk (Turkish dry fermented sausage). Tarragon was used in sucuk samples at three distinct levels (0.25%, 0.50% and 0.75%), and tarragon-free samples were used as control. Analyses were performed on the 0., 3., 5., 7., 9. and 13. days of ripening time. Although the presence of tarragon did not affect L* and b* values, the pH, a* (P<0.01), aw (P<0.05), and TBARS values of the sucuk samples were statistically affected. Tarragon demonstrated antioxidant activity in sucuk in accordance with 90 % +/- 0.3 DPPH scavenging activity. When the volatile profile was evaluated, the key component affected by the addition of tarragon was determined as Anisole, p-allyl-. As a result, while an antioxidative effect was detected during ripening in the presence of tarragon in sausage, undesirable changes were not observed on some quality characteristics such as pH and a(w).
- MakaleAçık Erişim2021Estimation of shear strength parameters of a high plasticity clayey soil stabilized with lime at different curing temperatures using Response Surface Methodology (RSM)
Kamiloglu, Hakan Alper; Turan, Huseyin
Response Surface Methodology (RSM) is useful to find the optimum value of the response, to examine the relationship between independent variables and responses, or to shorten very complicated and timeconsuming numerical analyses. Although the RSM is a commonly used method in many branches of engineering, the use of this method is limited in geotechnical studies. In this study, the effect of curing temperature on shear strength and unconfined compressive strength parameters of lime stabilized clayey soil was examined. Within the scope of the study, it was aimed to evaluate the Response Surface Method to estimate these parameters with less experimental effort. In the study effect of the temperature was controlled with five different curing temperature level (3 degrees C, 13 degrees C, 23 degrees C, 33 degrees C, 43 degrees C). Three axial (UU) and unconfined compression tests were performed on 40 pieces of stabilized soil samples cured along 28 days at various temperatures. The experimental design was performed with Response Surface Methodology to reduce the test number. Within this scope, an equation was derived statistically to estimate the deviator strength of the samples. Statistical analyses were performed considering deviator strengths of the 18 pieces of samples. As a result of the study, it was seen that it is possible to estimate shear strength and unconfined compression strengths of the samples accurately by using fewer test numbers with Response Surface Methodology.
AtıfWoS2OpenAlex2İncele - MakaleAçık Erişim2018Effect of the polynaphtalene sulfonate based superplasticizer on mechanical and physical properties of blended cement replaced with Bayburt Stone at different fineness
Tekin, Ilker; Tekin, Duygu
In this study, grinded Bayburt Stone has pozzolanic property with 45 mu m and 125 mu m fineness was replaced with PC 42.5R (CEM-I) type cement at ratio of 0%, 10%, 20%, and 30% by weight. The aim of this study was made up in two phase, one of them was to investigate the effects of Bayburt Stone waste contain zeolite with different fineness on physical and mechanical properties of composite cements and the other one was to investigate behavior of same superplasticizer belong to two different company. Standard consistency, setting time and volume expansion tests are performed on the cement pastes and compression strength and flow tests are performed on the cement mortars. Mortar specimens are kept in standard curing for 28 days, than compressive strength test are performed at the ages of 2, 7 and 28 days. As a result, changes in specific surface area depending on the particle size distribution of mineral additives makes significant different on the physical, mechanical and rheological properties of cement mortars. In this study, it was determined that same type of superplasticizers which had different companies could be used in blended cement replaced with Bayburt stone.
AtıfOpenAlex1TR Dizin1İncele - MakaleAçık Erişim2015INVESTIGATION OF RELIABILITIES OF BOLT DISTANCES FOR BOLTED STRUCTURAL STEEL CONNECTIONS BY MONTE CARLO SIMULATION METHOD
Öztekin, Ertekin
Design of the distance of bolts to each other and design of the distance of bolts to the edge of connection plates are made based on minimum and maximum boundary values proposed by structural codes. In this study, reliabilities of those distances were investigated. For this purpose, loading types, bolt types and plate thicknesses were taken as variable parameters. Monte Carlo Simulation (MCS) method was used in the reliability computations performed for all combination of those parameters. At the end of study, all reliability index values for all those distances were presented in graphics and tables. Results obtained from this study compared with the values proposed by some structural codes and finally some evaluations were made about those comparisons. Finally, It was emphasized in the end of study that, it would be incorrect of the usage of the same bolt distances in the both traditional designs and the higher reliability level designs.
AtıfWoS1OpenAlex3İncele - MakaleAçık Erişim2014DETERMINATION OF THE WEAR FAULT IN SPUR GEAR SYSTEM USING STATISTICAL PROCESS CONTROL METHOD
Maras, Sinan; Arslan, Hakan
Vibration analysis is one of the early warning methods widely used in obtaining information about faults occurring on the machine elements and structures. In this method, gear fault detection can be performed by analyzing of the vibration test results using signal processing, artificial intelligence and statistical analysis methods. The objective of this study is detection the existence of wear by examining changes in the vibrations of spur gears due to wear faults in statistical process control carts. In this study, artificial wear was created on the surfaces of spur gears in order to be examined in gears test rig. Then, these gears were attached and vibrations data were recorded by operating the system at various loading and number of cycles conditions. Detection of fault was demonstrated by analyzing undeformed and worn gears data in statistical process control carts by means of real-time experimental studies.
AtıfWoS1OpenAlex2İncele - MakaleYalnız künye2009The Comparison of Layout Arrangements for the Material Flow Ordering Planning in Production Systems through Simulation Analysis
Aksarayli, Mehmet; Altuntas, Serkan
Enterprises have to make suitable location planning to decrease their product cost and to increase their productivity in our time. The aim of this study is to compare the basic layout types used in arranging inside facility layout with each other by using simulation. Especially in this study, material handling times between machines and ratio of these times in total times were interested. First of all a new production system is designed to the basic layout types used in arranging inside facility layout. And then in the designed production system, machines are arranged for each machine layout types. Machine layout types are transferred to PROMODEL simulation software. Then with the results of analysis, material handling times of different machine layout types and ratio of material handling times in total production time were compared and the results obtained from this analysis were given after commented on this study.
AtıfWoS7İncele - DergiPAMUKKALE UNIVERSITY JOURNAL OF ENGINEERING SCIENCES-PAMUKKALE UNIVERSITESI MUHENDISLIK BILIMLERI DERGISI