Yayınlar
Koleksiyondaki öğeler
- MakaleAçık Erişim2022Scopus SJRQ1Zircon U[sbnd]Pb geochronology and petrology of the tholeiitic gabbro from the Kovanlık (Giresun) area: Constraints for the Late Cretaceous bimodal arc magmatism in the Eastern Pontides Orogenic Belt, NE Turkey
Temizel, İrfan; Arslan, Mehmet; Abdioğlu Yazar, Emel; Aslan, Zafer; Kaygusuz, Abdullah; Baki Eraydın, Tuğba
Mafic intrusions of the Eastern Pontides Orogenic Belt (EPOB) in NE Turkey are found rarely in varying ages, sizes and compositions. Of these, tholeiitic gabbro stocks, namely Kovanlık gabbro, consist of pyroxene gabbro and minor hornblende gabbro with a mainly granular texture. The pyroxene gabbro contains plagioclase, clino- and ortho-pyroxene, olivine, and Fe[sbnd]Ti oxides, whereas the hornblende gabbro includes mainly plagioclase, hornblende, rarely biotite and Fe[sbnd]Ti oxides. The clinopyroxene compositions of the pyroxene gabbro exhibit arc-related tholeiitic magmas. The Kovanlık gabbroic rocks have high U/Yb zircon ratios, implying continental arc-related tectonic settings. Zircon U[sbnd]Pb dating on the Kovanlık gabbro yielded a concordia age of ~92 Ma (Turonian). Petrochemically, the gabbroic rocks have a tholeiitic character and show arc- and MORB-like signatures. The gabbroic rocks show positive Rb, Ba, K, Pb, Sr, Ti and negative Nb, Ce, Nd, Hf, Zr anomalies in N-MORB normalized diagrams. The chondrite-normalized REE plots of the rocks show flat to feeble right-dipping REE patterns (LaN/LuN = 0.94–2.56) with positive Eu anomalies (EuN/Eu* = 1.15–2.14; except for two pyroxene gabbro samples). The gabbroic rocks have 87Sr/86Sr(i) ratios (0.704533 to 0.706265), εNd(i) values (4.04–4.85), 206Pb/204Pb (18.56–18.68) and 207Pb/204Pb (15.61–15.64) ratios. Low Nb/La and Nb/Y ratios, as well as high Sr/Yb and Ba/Rb ratios, indicate that source regions were modified by slab-derived fluids and/or melts. The obtained zircon dating, isotopic and geochemical data, as well as regional geology, revealed that the tholeiitic mafic magma derived from a spinel-lherzolite mantle at a relatively shallow depth, and the Kovanlık gabbro in the EPOB represents one of the first mafic intrusions of Late Cretaceous bimodal arc magmatism related to the roll-back of the northwards subducting Tethys oceanic plate.
AtıfScopus19WoS17OpenAlex20
İncele - MakaleAçık Erişim2023Scopus SJRQ12D and 3D Numerical Simulation of Dam-Break Flooding: A Case Study of the Tuzluca Dam, Turkey
Akgun, Cagri; Nas, Salim Serkan; Uslu, Akin
Unlike river floods, floods caused by dam breaks occur much more abruptly and at higher water velocities. The failure of a dam due to any reason can pose a significant threat to settlements in the downstream area due to the high volume of water in its reservoir. In this case study, 2D and 3D numerical failure simulations of Tuzluca Dam, to be built in the Igdir province of Turkey, was performed using the Flow3D Hydro software. The Shallow Water and Reynolds-Averaged Navier-Stokes (RANS) equations were used to simulate flood propagation. In the numerical models, RNG k-ε was chosen as the turbulence model. In the analyses carried out under different scenarios, it was observed that in both methods (2D and 3D), significant flood depths would occur in two downstream village settlements. Within the scope of the study, the mortality rates in different dam-break scenarios of the Tuzluca Dam were also investigated. It has been observed that the dam failure time and breach geometry have a significant impact on the flood propagation and the loss of life in settlements. Additionally, a dam-break experiment in literature dam-break experiment was numerically solved using the Flow3D Hydro software to validate the numerical model, and a satisfactory harmony was observed between the experimental results and the numerical models. Although the 3D analysis provided a more accurate representation of the experimental results, it took approximately 20 times longer to complete the dam failure simulation in the case study compared to the 2D model. As a result, it has been determined that for comprehensive studies such as those of dam failures, 2D analyses may be more practical, while for smaller-scale scenarios or situations requiring greater precision, 3D modelling is appropriate.
AtıfScopus14WoS12OpenAlex12İncele - MakaleAçık Erişim2024Scopus SJRQ1Optical Filter Design for Multi-Color VLC in Reflective Environments with Multiple Lighting Sources
Kucuk, Mehmet A.; Albayrak, Cenk; Turk, Kadir; Yanikomeroglu, Halim
Multi-color visible light communication (MC-VLC) systems, which aim to transmit data in parallel with multi-color LEDs, use optical filters to separate color channels at the receiver side. However, the relatively wide spectrum of LEDs may cause inter-color interference (ICI). In addition, the passband of the optical filters shifts towards the blue wavelength depending on the angle of incidence (AoI) of the light, which may also increase ICI. In this paper, the problem of designing optical filters for MC-VLC systems in the presence of the blue-shift is investigated. The communication environment is assumed to be a typical room illuminated by multiple fixtures in accordance with lighting standards. Moreover, the walls in the room are reflective, resulting in non-line-of-sight (NLoS) components. The passband edges of the optical filter are optimized to maximize the average signal-to-interference-plus-noise ratio (SINR). The effect of the field-of-view (FOV) of the receiver on the system performance is also analyzed. Finally, a generic optical filter has been designed that can be used in standards-compliant environments for selected LEDs and a given FOV range. The results show that the designed optical filter provides a higher and uniform capacity than those in the literature.
AtıfScopus6WoS3İncele - MakaleYalnız künye2024Scopus SJRQ1Spatiotemporal analysis of transition probabilities of wet and dry days under SSPs scenarios in the semi-arid Susurluk Basin, Türkiye
Şan, Murat; Nacar, Sinan; Kankal, Murat; Bayram, Adem
Precipitation, especially in regions dominated by the Mediterranean climate, is one of the most critical parameters of the hydrological cycle and the environment affected by climate change. One the one hand, the transition probabilities of wet and dry days in precipitation occurrence are a relatively new topic, on the other hand these are essential in defining the regional climate. For the first time, spatiotemporal variations of transition probabilities of wet and dry days in the Susurluk Basin, northwestern Türkiye, dominated by a semi-arid Mediterranean climate and also having a mountain climate, were analyzed based on the observation (1979–2014) and future terms (2030–2059 as short and 2070–2099 as long), under four Shared Socioeconomic Pathways (SSPs) scenarios. To do this, statistical downscaling was performed for 14 general circulation models (GCMs) from the CMIP6. By applying an ensemble of four high-performing GCMs, four indices for transition probabilities of wet and dry, i.e., a dry day following a dry day (FDD), a wet day following a dry day (FDW), a dry day following a wet day (FWD), and a wet day following a wet day (FWW), were calculated, and their changes were determined statistically. Monotonic and partial trends of the indices were also analyzed. According to the results, the FDD will increase in water year and wet period and autumn in the future, especially for the long term, in the basin dominated by the FDD (75 % in water year). The risks are higher in the western part of the basin, where human activities are intense, as the FDD is higher in this part than other parts especially in summer (90–100 %) in SSP3-7.0 and SSP5-8.5 scenarios for the long term. So, the length of consecutive dry days in the wet period and water year will increase in the basin, thus increasing the likelihood of droughts. As for the intra-term trends, the FDD increases and the FWW decreases in the water year and seasons in SSP3-7.0 and SSP5-8.5, contrary to the observation term.
AtıfScopus20WoS19OpenAlex20
İncele - MakaleAçık Erişim2023Scopus SJRQ3The epiphytic bryophyte succession of Buxus sempervirens forests in the Fırtına Valley, Rize (North Türkiye)
Ezer, Tülay; Alataş, Mevlüt; Batan, Nevzat; Erata, Hüseyin
In this study, the epiphytic bryophyte succession of the Buxus sempervirens L. forests in Fırtına Valley (Çamlıhemşin-Rize, North Türkiye), one of the nine biodiversity hotspots in Türkiye, was investigated. For this pur-pose, a total of 60 sampling plots were taken from the live trunks of B. sempervirens trees of different ages. Twenty-nine epiphytic bryophyte species were determined (24 mosses and 5 liverworts) within the sample plots. Also, six different life form types and four different habitat affinity categories were determined. Among them, the mat type life form is in first place with 34.4% and the cortico-saxicolous species are the most common with 51.7%. Two-way indicator species analysis (TWINSPAN) classified the epiphytic bryophyte communities on B. sempervirens trunks at the second level into two main clusters (A and B) and three sub-clusters (A1, B1 and, B2). Detrended correspondence analysis (DCA) axis 1 was interpreted as gradient along the height of the epiphytic habitat (from the lower base to the upper zone) on trunks and the DCA axis 2 was interpreted as gradient of moisture (from mesic to xeric). Exsertotheca crispa (Hedw.) S. Olsson, Enroth & D. Quandt was the species with the highest index of ecological significance (IES) value on the lower bases of the aged trees. Species diversity and epiphytic cover in the upper zones were lower than in the basal and middle zones in the study area. While Metzgeria furcata, (L.) Corda, Oxyrrhynchium hians (Hedw.) Loeske, Plagiothecium nemorale (Mitt.) A.Jaeger, and Radula lindenbergiana Gottsche ex C.Hartm were only found on old trees, Ctenidium molluscum (Hedw.) Mitt. and Pseudoleskeella nervosa (Brid.) Nyholm were only found on middle-aged trees.
AtıfScopus1WoS1OpenAlex1İncele - MakaleAçık Erişim2023Experimenting and thematically analyzing the project study to increase effectiveness of the case of the formation of seasons
Semercioğlu, Melike Güzin; Kalkan, Hüseyin
Teacher education is a critical component of the educational system, as it provides the skills and knowledge necessary for effective instruction in science and other related subjects. The purpose of this research is to enhance the incomplete and incorrect knowledge and skills of science teachers related to the formation of seasons by using a physical model that reveals the “change in the amount of energy per unit surface (CAEUS)”. Twenty-nine teachers from different regions of the country conducted experimental studies with physical models related to the formation of seasons, as part of a 5-day training program that included topics on astronomy. In this study, in which descriptive statistics, inferential statistics, and thematic analysis were used, it was observed that the comprehension of the causes and effects of the change in the amount of energy per unit surface was very successful in the development of mental models for understanding the seasons. Furthermore, the teachers reported that they were able to use the knowledge and skills they gained from the program in their teaching practice. The results obtained were compared with the literature and interpreted accordingly.
- KitapYalnız künye2023Spirituality management in the workplace: New strategies and approaches
Özsungur, Fahri; Bekar, Fevziye
In recent years, after realizing that personal beliefs and spiritual development are as important as mental strength, studies in the field of spirituality in the workplace have grown exponentially. The experts here provide conceptual frameworks and guidance by examining the subject in the light of current developments at multiple levels of analysis: individual, organizational, cultural, and in leadership. Furthermore, this book focuses on rapidly evolving business models: remote working, the cyber-workplace, social media, digitalization, etc - all accelerated by the COVID-19 epidemic. The concept of spirituality in the workplace can be harmful as well as beneficial. Employees who are spiritually attached to the workplace may ignore issues known as 'organizational deviation' (such as theft and corruption) by keeping the interests of the institution in the foreground. These 'dark' and 'invisible' aspects of spirituality in the workplace are also examined, with a special focus on identifying aspects of spirituality which can harm businesses. Spirituality in the workplace considers employees as a whole, in spirit, body, and mind.
- MakaleAçık Erişim2024Scopus SJRQ1A robust chaos-inspired artificial intelligence model for dealing with nonlinear dynamics in wind speed forecasting
Barış, Caner; Yanarateş, Cağfer; Altan, Aytaç
The global impacts of climate change have become increasingly pronounced in recent years due to the rise in greenhouse gas emissions from fossil fuels. This trend threatens water resources, ecological balance, and could lead to desertification and drought. To address these challenges, reducing fossil fuel consumption and embracing renewable energy sources is crucial. Among these, wind energy stands out as a clean and renewable source garnering more attention each day. However, the variable and unpredictable nature of wind speed presents a challenge to integrating wind energy into the electricity grid. Accurate wind speed forecasting is essential to overcome these obstacles and optimize wind energy usage. This study focuses on developing a robust wind speed forecasting model capable of handling non-linear dynamics to minimize losses and improve wind energy efficiency. Wind speed data from the Bandırma meteorological station in the Marmara region of Turkey, known for its wind energy potential, was decomposed into intrinsic mode functions (IMFs) using robust empirical mode decomposition (REMD). The extracted IMFs were then fed into a long short-term memory (LSTM) architecture whose parameters were estimated using the African vultures optimization (AVO) algorithm based on tent chaotic mapping. This approach aimed to build a highly accurate wind speed forecasting model. The performance of the proposed optimization algorithm in improving the model parameters was compared with that of the chaotic particle swarm optimization (CPSO) algorithm. Finally, the study highlights the potential of utilizing advanced optimization techniques and deep learning models to improve wind speed forecasting, ultimately contributing to more efficient and sustainable wind energy generation. This robust hybrid model represents a significant step forward in wind energy research and its practical applications.
AtıfScopus58WoS55OpenAlex55İncele - MakaleAçık Erişim2023Scopus SJRQ4Detection of pneumonia from pediatric chest X-ray images by transfer learning
Demir, Yasin; Bingöl, Özkan
When pathogens such as viruses, bacteria and fungi attack the lungs, the alveoli fill with inflamed fluid, causing pneumonia. Early diagnosis of this disease, which has fatal outcomes especially in children under 5 years old, is very important in controlling undesirable situations. Chest X-ray images play an important role in the diagnosis of pneumonia. In addition, the fact that the amount of radiation is lower than imaging devices such as tomography and the possibility of being accessible even from rural areas creates an advantage for X-ray devices. However, X-ray images that are not always clear or human conditions such as fatigue and lack of attention can make it difficult for specialists to detect pneumonia. In this study, a transfer learning-based convolutional neural network (CNN) approach is proposed, which can help specialists in the early and accurate diagnosis of pneumonia in children and, classify healthy and diseased individuals through Chest X-ray images. As a result of the study, an original CNN was proposed by adding additional layers to the AlexNet architecture layers and a test accuracy of 96.31% was obtained. Copyright 2021, Yıldız Technical University.
AtıfScopus9WoS4İncele - MakaleAçık Erişim2024Scopus SJRQ2Effects of Ultrafine Fly Ash against Sulphate Reaction in Concrete Structures
Demir Şahin, Demet; Eker, Hasan
In this study, Afşin Elbistan C-type fly ash (FA) was used, which protects against the sulphate reaction that damages concrete. The detrimental effects of post-reaction decrease with increasing fly ash fineness. The study used 10%, 30%, and 50% weight substitutes of cement. The fly ash was ground in a ball mill for 0, 10, 20, 30, 45, and 60 min, and Blaine fineness values of 1555, 1632, 2347, 2589, 2766, and 3433 cm2/g were obtained, respectively. The effect of the samples on the sulphate resistance was investigated by exposing the samples to 5% or 10% added sulphate solutions, and the compressive strength and ultrasonic pulse velocity of the concrete were tested. The compressive strength values obtained decreased with the increase in sulphate content, and the increase in the grinding time and the amount of substituted FA increased the compressive strength values. It was observed that weight loss increased with increasing sulphate content and decreased with the addition of FA with a high Blaine fineness. It was determined that as the Blaine fineness value increased, the sulphate content, FA substitution amount, and ultrasonic pulse speed decreased. This study was carried out to determine the effects of fly ash used at different fineness and replacement ratios on the performance and strength of concrete after exposure to external influences such as sulphate. The use of fly ash instead of cement will reduce the use of waste materials and natural resources and prevent environmental pollution. The cost of cement and concrete will be reduced.
AtıfScopus19WoS18OpenAlex23İncele - MakaleYalnız künye2023Institutional Quality and Unemployment in OECD Nations: A Panel Causality Test Using a Fourier Function
Cicen, Yildirim Beyazit
A significant impact of institutional quality on sustainable growth has been revealed in recent studies. This study aims to examine the relationship between different institutions and their unemployment rates, considering the legal system and property rights index. A novel and robust panel Fourier causality test was conducted to investigate 38 OECD nations for the relationship between their institutional quality and unemployment rates between the years 2000 and 2018. Regardless of stationarity, the adopted method allowed for the endogenous identification of structures, locations, and forms, and the evaluation of causal relationships. In addition, the Fourier method and the trigonometric terms were used for analyzing the structural changes. According to the findings, a unidirectional causal relationship existed between the quality of the legal system and property rights and the unemployment rates in the investigated nations during the relevant period. In addition, bidirectional causality existed between the variables in Germany and the United States. It is therefore recommended to necessarily improve a nation’s institutional structure to reduce its unemployment rate. Low unemployment rates would be achieved by improving institutional quality, particularly in developing nations. Therefore, policymakers must focus on formulating policies to reduce unemployment in the long run, considering the legal system and property rights of the nation.
AtıfScopus3WoS2İncele - MakaleYalnız künye2024Scopus SJRQ2Dual Capacitive Behavior of Hollow Porous Gold Nanoparticle/Multi-Walled Carbon Nanotube Composite Films as Anode and Cathode Electrode Materials for Supercapacitors
Bahar, Necmiye; Ekinci, Duygu
In this paper, composites of multi-walled carbon nanotubes decorated with hollow porous gold nanoparticles (HPAuNPs/MWCNTs) were prepared using wet chemistry methods, and their capacitive properties were evaluated in Na2SO4 aqueous solution on glassy carbon (GC) electrodes as both cathode and anode materials by cyclic voltammetry (CV), galvanostatic charge–discharge (GCD), and electrochemical impedance spectroscopy (EIS) techniques. Also, the MWCNTs were studied under similar experimental conditions for comparison. The results revealed that the charge storage mechanisms of the MWCNTs/GC and HPAuNPs/MWCNTs/GC electrodes were electrostatic in the negative potential window, whereas they exhibited pseudocapacitive energy storage behavior in the positive potential region. Moreover, the HPAuNPs/MWCNTs/GC electrode yielded specific capacitance values of 57 and 49 F g−1 at a scan rate of 20 mV s−1 in the cathodic and anodic potential windows, respectively, which were higher than those of the MWCNTs/GC electrode (~9 F g−1). Graphic Abstract: [Figure not available: see fulltext.].
AtıfScopus6WoS6OpenAlex6İncele - MakaleAçık Erişim2023Scopus SJRQ3Geometric properties of timelike surfaces in Lorentz-Minkowski 3-space
Mazlum, Sümeyye Gür
In this paper, the relationships between geodesic torsions, normal curvatures and geodesic curvatures of the parameter curves intersecting at any angle on timelike surfaces in Lorentz-Minkowski 3-space are obtained by various equations. In addition, new equivalents of well-known formulas (O. Bonnet, Euler, Liouville) are found in this space. Finally, the examples of these surfaces are given.
AtıfScopus16WoS16OpenAlex16İncele - MakaleAçık Erişim2023Scopus SJRQ3Exploring managerial insights through multi criteria decision making techniques in pharmacy inventory classification problem
Şimşek, Ahmet Bahadır; Göktekin, Zekiye; Geliç, Büşra
The current research addresses the inventory classification problem of community pharmacies, which have a dual role as both a vital component of the pharmaceutical supply chain and a typical retail store. Despite the existing literature indicating that pharmacists may lack knowledge on inventory management, it seems that the MCIC literature is weak in explaining how pharmacists can benefit from MCDA techniques in all aspects. To bridge this gap, the study aims to demonstrate that pharmacists can utilize MCDA techniques to gain deeper insights beyond mere classification in the context of inventory management. Real-world data from a community pharmacy in Turkey was classified using the EDAS method. Sensitivity analysis was performed for MCDA inputs, about which pharmacists may lack information. Scenario findings based on criterion weights and threshold values offer important managerial implications for pharmacists. This study provides a critical contribution to the literature on inventory management in community pharmacies by highlighting the potential of MCDA techniques to support decision-making beyond mere classification. The sensitivity analysis also sheds light on areas where pharmacists may lack knowledge and suggests ways to address these gaps. Overall, the study underscores the need for pharmacists to have a deeper understanding of inventory management and highlights the potential benefits of MCDA techniques in addressing this challenge.
AtıfScopus4WoS1OpenAlex6İncele - MakaleYalnız künye2024Scopus SJRQ1Perspectives on the mental health of immigrant children in the pediatric emergency department: a phenomenological study
Uzun, Sevda; Ustuner Top, Fadime
The concept of immigration is a complex phenomenon that goes far beyond mere relocation and refers to the transition from one social environment to another or from one culture to another. Immigration and its effects on children are regarded as global and interdisciplinary problems. This study aims to reveal the opinions of nurses working in the pediatric emergency department about the mental health of immigrant children. In this qualitative research, a phenomenological research design was used (N = 19).Interviews were conducted using a semi-structured interview form. Colaizzi’s method was used to analyze the data manually. Four categories, eleven themes, and thirty-eight sub-themes emerged from the analysis of the data. The difficulties nurses experience with immigrant children and their relatives in providing care include communication problems, cultural problems, and economic problems. The causes of mental health problems in immigrant children include war, exposure to immigration traumas, cultural assimilation problems, loss/absence of parents, economic reasons/excessive role overload, and exposure to neglect and/or abuse. Education, economic aid, communication, non-stigmatization, and social support were identified as significant sub-themes for practices to protect/promote children’s mental health. The study showed that immigrant children experience physical and mental problems. Nurses struggle to treat immigrant children effectively but have difficulties due to insufficient facilities. It may be suggested to include cultural care models in patient care.
- MakaleYalnız künye2022Scopus SJRQ1A decision support tool for the student–supervisor allocation problem of postgraduate programs
Bahadır Şimşek, Ahmet
The stable supervisor-student allocation (SSA) is crucial in the success of the research at the postgraduate education. The difference of SSA from the well-known student-project allocation (SPA) problem is that it moves the determination of the research topic to after the allocation, thus allowing the student and the lecturer to discuss the research topic. In the education systems of developing countries where decision support systems are not prevalent, the student–lecturer allocation is performed manually with a greedy approach, which can result in unstable matches. In this study, we propose a multi-objective binary model that allows decision-makers to shift between three main allocation strategies with a trade-off parameter. We test the model with a real dataset of a postgraduate program in Turkey. Experiments show that the model can successfully implement different strategies, and there can also be more than one breaking point between strategies. Also, we transform the model into a decision support tool that provides easy use to the decision-maker and reports the solutions of the trade-off parameter sensitivity analysis. The tool reports the level of satisfaction for each strategy at different breakpoints, allowing the decision-maker to see its options.
AtıfScopus6WoS3OpenAlex5İncele - MakaleAçık Erişim2023Scopus SJRQ4The Effect of Different Physical Education Class Hours Applied in Secondary Education on Certain Physiological and Physical Parameters
Sever, Mehmet Onur; Söyler, Mehmet; Çingöz, Yunus Emre; Barkin, Emrah
This study investigates the effects of different physical education lesson hours applied to secondary school students on some physical and physiological parameters. A total of 36 secondary school students voluntarily participated in the study: respectively, 18 from Group 1 and 18 from Group 2 (age: 14.94±0.41 / 14±0.34 years; height: 175.22±7.62 / 169.33±6.95 cm; weight: 61.95±9.07 / 56.83±6.33 kg). While Group 1 students received 8 hours of physical education lessons per week, Group 2 students were given 2 hours of physical education lessons according to the Ministry of National Education curriculum. Body weight (kg) and body fat percentage (%) were determined using the Inbody 270 professional body analysis device. A portable dynamic balance device was used to determine the students' balance. The test results were analyzed using the SPSS Statistics software. Non-parametric tests were employed since the data did not show a normal distribution. The Mann-Whitney U test was used to test the significance of the difference in balance and body fat percentage (BFP) level averages between the two independent groups. Upon evaluating our data, it was statistically ascertained that the dynamic balance (Balance) values of Group 1 secondary school students were significantly higher when compared to Group 2 students. However, no statistically significant difference was found in body fat percentage (BFP) values between Group 1 and Group 2. As a result, it is seen that increasing physical education lesson hours in secondary education contributes to some physical development of young individuals.
- MakaleAçık Erişim2023Scopus SJRQ1Machine learning and domain adaptation to monitor yoghurt fermentation using ultrasonic measurements
Bowler, Alexander; Ozturk, Samet; di Bari, Vincenzo; Glover, Zachary J.; Watson, Nicholas J.
In manufacturing environments, real-time monitoring of yoghurt fermentation is required to maintain an optimal production schedule, ensure product quality, and prevent the growth of pathogenic bacteria. Ultrasonic sensors combined with machine learning models offer the potential for non-invasive process monitoring. However, methods are required to ensure the models are robust to changing ultrasonic measurement distributions as a result of changing process conditions. As it is unknown when these changes in distribution will occur, domain adaptation methods are needed that can be applied to newly acquired data in real-time. In this work, yoghurt fermentation processes are monitored using non-invasive ultrasonic sensors. Furthermore, a transmission based method is compared to an industrially-relevant non-transmission method which does not require the sound wave to travel through the fermenting yoghurt. Three machine learning algorithms were investigated including fully-connected neural networks, fully-connected neural networks with long short-term memory layers, and convolutional neural networks with long short-term memory layers. Three real-time domain adaptation strategies were also evaluated, namely; feature alignment, prediction alignment, and feature removal. The most accurate method (mean squared error of 0.008 to predict pH during fermentation) was non-transmission based and used convolutional neural networks with long short-term memory layers, and a combination of all three domain adaption methods.
AtıfScopus28WoS26OpenAlex33İncele - MakaleAçık Erişim2023Scopus SJRQ1Near-infrared spectroscopy and machine learning for classification of food powders during a continuous process
Ozturk, Samet; Bowler, Alexander; Rady, Ahmed; Watson, Nicholas J.
In food production environments, the wrong powder material is occasionally loaded onto a production line which impacts food safety, product quality, and production economics. The aim of this study was to assess the potential of using Near Infrared (NIR) spectroscopy combined with Machine Learning to classify food powders under motion conditions. Two NIR sensors with different wavelength ranges were compared and the ML models were tasked with classifying between 25 food powder materials. Eleven different spectra pre-processing methods, three feature selection methods, and five algorithms were investigated to find the optimal ML pipeline. It was found that pre-processing the spectra using autoencoders followed by using support vector machines with the all spectral wavelengths from both sensors was most accurate. The results were improved further using under-sampling and boosting. Overall, this method achieved 99.52, 97.12, 94.08, and 91.68% accuracy for the static, 0.017, 0.036 and 0.068 m s-1 sample speeds. The models were also validated using an independent test sets.
AtıfScopus85WoS69OpenAlex75İncele - MakaleAçık Erişim2022Scopus SJRQ1COVID-DSNet: A novel deep convolutional neural network for detection of coronavirus (SARS-CoV-2) cases from CT and Chest X-Ray images
Reis, Hatice Catal; Turk, Veysel
COVID-19 (SARS-CoV-2), which causes acute respiratory syndrome, is a contagious and deadly disease that has devastating effects on society and human life. COVID-19 can cause serious complications, especially in patients with pre-existing chronic health problems such as diabetes, hypertension, lung cancer, weakened immune systems, and the elderly. The most critical step in the fight against COVID-19 is the rapid diagnosis of infected patients. Computed Tomography (CT), chest X-ray (CXR), and RT-PCR diagnostic kits are frequently used to diagnose the disease. However, due to difficulties such as the inadequacy of RT-PCR test kits and false negative (FN) results in the early stages of the disease, the time-consuming examination of medical images obtained from CT and CXR imaging techniques by specialists/doctors, and the increasing workload on specialists, it is challenging to detect COVID-19. Therefore, researchers have suggested searching for new methods in COVID- 19 detection. In analysis studies with CT and CXR radiography images, it was determined that COVID-19-infected patients experienced abnormalities related to COVID-19. The anomalies observed here are the primary motivation for artificial intelligence researchers to develop COVID-19 detection applications with deep convolutional neural networks. Here, convolutional neural network-based deep learning algorithms from artificial intelligence technologies with high discrimination capabilities can be considered as an alternative approach in the disease detection process. This study proposes a deep convolutional neural network, COVID-DSNet, to diagnose typical pneumonia (bacterial, viral) and COVID-19 diseases from CT, CXR, hybrid CT + CXR images. In the multi-classification study with the CT dataset, 97.60 % accuracy and 97.60 % sensitivity values were obtained from the COVID-DSNet model, and 100 %, 96.30 %, and 96.58 % sensitivity values were obtained in the detection of typical, common pneumonia and COVID-19, respectively. The proposed model is an economical, practical deep learning network that data scientists can benefit from and develop. Although it is not a definitive solution in disease diagnosis, it may help experts as it produces successful results in detecting pneumonia and COVID-19.
AtıfScopus36WoS22OpenAlex36İncele
