Sürdürülebilir Kalkınma Amaçları

• 2
• 1
• 1

• MakaleAçık Erişim2025Scopus SJRQ1/r/munzur/1290
  Hybrid GA-ConvLSTM for data-driven prediction of climate variables: a case study of the most biodiverse cities in India

  Utku, Anıl; Akyol, Sinem

  Climate variables affect ecosystems’ structure, sustainability, and biodiversity. Monitoring and analysis of climate variables are essential for the management of ecosystems and the conservation of biodiversity. Temperature affects the distributions, habitats, and phonological events of living things in ecosystems. Humidity is essential for meeting plants’ water needs and the habitat’s quality and health. Dew point is essential for water quality, balance, plant health, and microclimate. Wind speed is essential for pollinating plants, microhabitats, and the health of living things. For such reasons, accurate prediction of climate variables using artificial intelligence methods can effectively predict the future of ecosystems and biodiversity and develop future strategic plans. This study aimed to predict the climate variables of temperature, humidity, dew point, and wind speed of Bhubaneshwar, Dehradun, Guwahati, Mysore, and Nagpur, among the cities with the highest biodiversity in India. It aimed to determine the long-term impacts of climate change on managing ecosystems and biodiversity protection. The hyperparameters of the ConvLSTM model, developed using CNN and LSTM, were optimized with a genetic algorithm, and the GA-ConvLSTM hybrid model was created. GA-ConvLSTM was tested with XGBoost, SVM, ANFIS, CNN, LSTM, and base ConvLSTM using approximately 15 years of hourly, up-to-date, and real-time climate variable data of cities. This work is one of the pioneering studies using GA to optimize deep learning hybrid models for multi-city climate prediction through extensive long-term high-resolution datasets. The experimental results reveal that GA-ConvLSTM outperforms conventional models, producing an R2 score exceeding 0.9 for each city across all climate variables. GA-ConvLSTM produced temperature predictions with 0.833 RMSE and 0.995 R2 for Bhubaneswar, humidity predictions with 3.461 RMSE and 0.990 R2, dew point predictions with 0.632 RMSE and 0.995 R2, and wind speed predictions with 2.278 RMSE and 0.993 R2. The experimental findings confirm the improved performance over conventional models due to SVM and XGBoost failing to achieve these R2 values and high RMSE scores. The research meets three principal obstacles because of the combination of stochastic climate datasets and complex nonlinear patterns, which also encounter limitations in conventional methods for managing scale-dependency relationships and manual parameter adjustments. This approach delivers three main strengths because it combines CNN and LSTM networks through genetic programming while tracking temporal and spatial relationships and processing noisy data automatically, leading to improved results compared to other methods. Unlike other published models in the field, the hybrid model GA-ConvLSTM combines optimized CNN and LSTM elements with genetic algorithm elements. The proposed model received its first application in this research for multi-city climate prediction through training with 15 years of hourly data. Standard methods encounter challenges when detecting long-term dependencies, but GA-ConvLSTM successfully addresses this issue.

  • 
  AtıfScopus1·OpenAlex1

  İncele
• MakaleAçık Erişim2025Scopus SJRQ1/r/munzur/710
  New comparative approach to multi-level thresholding: chaotically initialized adaptive meta-heuristic optimization methods

  Serbet, Fatmanur; Kaya, Turgay

  One method aimed at enhancing the performance of meta-heuristic optimization techniques is the incorporation of chaotic systems. Instead of irregular distributions in the search space, chaotic distributions are employed in the initial population of optimization algorithms to improve the efficiency of the search process. This approach enables search agents distributed in a chaotic manner to effectively explore the search space. The initial populations of both the well-established PSO algorithm and the enhanced WSO algorithm, which incorporates advanced search techniques, are distributed in the search space according to the characteristics of Logistic, Chebyshev, Circle, Sine, and Piecewise chaotic maps in this study. The original PSO and WSO algorithms, as well as the resulting chaotically initialized PSO and chaotically initialized WSO algorithms, were tested using 23 benchmark functions. Subsequently, the Otsu method was integrated into the tested optimization algorithms to obtain multi-level thresholding values. These algorithms were applied to five different test images with a manually determined number of thresholds. The results obtained were presented in the study and evaluated using statistical tests.

  AtıfScopus13·OpenAlex12

  İncele
• MakaleAçık Erişim2024Scopus SJRQ1/r/munzur/2045
  Spread patterns of COVID-19 in European countries: hybrid deep learning model for prediction and transmission analysis

  Utku, Anıl; Akcayol, M. Ali

  The COVID-19 pandemic has profoundly impacted healthcare systems and economies worldwide, leading to the implementation of travel restrictions and social measures. Efforts such as vaccination campaigns, testing, and surveillance have played a crucial role in containing the spread of the virus and safeguarding public health. There needs to be more research exploring the transmission dynamics of COVID-19, particularly within European nations. Therefore, the primary objective of this research was to examine the spread patterns of COVID-19 across various European countries. Doing so makes it possible to implement preventive measures, allocate resources, and optimize treatment strategies based on projected case and mortality rates. For this purpose, a hybrid prediction model combining CNN and LSTM models was developed. The performance of this hybrid model was compared against several other models, including CNN, k-NN, LR, LSTM, MLP, RF, SVM, and XGBoost. The empirical findings revealed that the CNN-LSTM hybrid model exhibited superior performance compared to alternative models in effectively predicting the transmission of COVID-19 within European nations. Furthermore, examining the peak of case and death dates provided insights into the dynamics of COVID-19 transmission among European countries. Chord diagrams were drawn to analyze the inter-country transmission patterns of COVID-19 over 5-day and 14-day intervals.

  • 
  AtıfScopus6·WoS3·OpenAlex7

  İncele
• MakaleAçık Erişim2024Scopus SJRQ1/r/munzur/2014
  A new hybrid approach for grapevine leaves recognition based on ESRGAN data augmentation and GASVM feature selection

  Doğan, Gürkan; Imak, Andaç; Ergen, Burhan; Sengur, Abdulkadir

  Grapevine leaf is a commodity that is collected only once a year and has a high return on investment due to its export. However, only certain types of grapevine leaves are consumed. Therefore, it is extremely important to distinguish the types of grapevine leaves. In particular, performing this process automatically on industrial machines will reduce human errors, workload, and thus cost. In this study, a new hybrid approach based on a convolutional neural network is proposed that can automatically distinguish the types of grapevine leaves. In the proposed approach, firstly, the overfitting of network models is prevented by applying data augmentation techniques. Second, new synthetic images were created with the ESRGAN technique to obtain detailed texture information. Third, the top blocks of the MobileNetV2 and VGG19 CNN models were replaced with the newly designed top block, effectively extracting features with the data. Fourthly, the GASVM algorithm was adapted and used to create a subset of the features to eliminate the ineffective and unimportant ones from the obtained features. Finally, SVM classification was performed with the feature subset consisting of 314 features, and approximately 2% higher accuracy and MCC score were obtained compared to the approaches in the literature.

  • 
  AtıfScopus8·WoS6·OpenAlex9

  İncele
• MakaleYalnız künye2024Scopus SJRQ1/r/munzur/1713
  A novel algorithm based on DNA coding for substitution box generation problem

  Artuğer, Fırat

  Substitution boxes are very important for cryptographic structures. But getting an 8-bit strong s-box is a difficult problem. To address this problem, a new approach is proposed in this study. DNA chains are used in the proposed approach. First of all, a weak s-box is obtained with the help of a chaotic map. Then, a DNA chain is formed with the same chaotic map. In this study, 2000-long DNA chains were produced. Then, two random values consisting of four components are determined from these DNA chains, and these values are addressed to the s-box structure. By changing the places of these values, the s-box structure is improved up to a certain iteration. In this study, 10,000 iterations were examined, and three different strong s-boxes were obtained by using three different DNA chains. The nonlinearity values of these s-box structures were observed as 110.25, 110.5, and 110.75, respectively. The proposed study has proven to work effectively with effective results in three different chains. In addition, it has been proven as a result of the analysis that the proposed s-box structures also have other cryptographic features. The proposed algorithm has surpassed many s-box generation philosophies in the literature. In addition, the proposed algorithm will provide serious gains to cryptographic structures that will be developed in the future.

  AtıfScopus21·WoS22·OpenAlex24

  İncele
• MakaleAçık Erişim2022Scopus SJRQ1/r/munzur/2731
  SBOX-CGA: substitution box generator based on chaos and genetic algorithm

  Artuğer, Fırat; Özkaynak, Fatih

  What makes artificial intelligence techniques so remarkable in the field of computer science is undoubtedly their success in producing effective solutions to difficult computational problems. In particular, metaheuristic optimization algorithms are a unique example of using artificial intelligence techniques to generate approximate solutions to problems that cannot be solved in polynomial time, called NP. Obtaining a substitution box (s-box) structure that will satisfy the desired requirements in cryptography is an example of these NP problems. In the literature, it is a hot topic to optimize the s-box structures obtained from chaotic entropy sources with heuristic algorithms to improve their cryptographic properties. The study with the highest nonlinearity value (110.25) based on optimization algorithms to date has been published in 2020. In this study, a method with a higher nonlinearity value than the algorithms previously proposed in the literature is developed. It has been shown that the nonlinearity value can be increased to 111.75. These results will be a basis for new research on the chaos-based s-box literature and will motivate new studies to develop alternative optimization algorithms in the future to obtain s-box structures based on the random selection equivalent to the AES s-box.

  AtıfScopus47·WoS35·OpenAlex49

  İncele
• MakaleAçık Erişim2021Scopus SJRQ1/r/munzur/2831
  An integrated interval type 2 fuzzy AHP and COPRAS-G methodologies for supplier selection in the era of Industry 4.0

  Kayapinar Kaya, Sema; Aycin, Ejder

  Supply chain has a very extensive and dynamic structure that incorporates new business models, new customer expectations, market searches and technological developments. With the introduction of Industry 4.0 into supply chain, a rapid and intensive process of digitalization begins to transform every step of supply chain. Supply chain selection is one of the essential decisions in reducing the supply chain cost and improving overall quality of product and services. With the implication of digital technologies and Industry 4.0 on supply chain, the supplier selection process has been significantly changed during the recent years. Companies are willing to need new requirements for their own suppliers in accordance with Industry 4.0 implementations and technologies. This paper aims to identify key criteria to Industry 4.0 technologies and evaluate them to select the right suppliers selection in the era of Industry 4.0. Within the scope of this study attempts to develop an integrated interval type 2 fuzzy AHP and GOPRAS-G methodology to select the appropriate supplier in the face of Industry 4.0 implementations. For this purpose, interval type 2 fuzzy AHP was employed to weight the supplier evaluation criteria and then, Gray COPRAS method has been applied to prioritize suppliers. This paper is to provide practitioners and researchers with insight into how Industry 4.0 strategies influence on supplier selection.

  AtıfScopus63·WoS51

  İncele
• MakaleYalnız künye2020Scopus SJRQ1/r/munzur/3825
  A deep convolutional neural network model for automated identification of abnormal EEG signals

  Yıldırım, Özal; Baloglu, Ulas Baran; Acharya, U. Rajendra

  Electroencephalogram (EEG) is widely used to monitor the brain activities. The manual examination of these signals by experts is strenuous and time consuming. Hence, machine learning techniques can be used to improve the accuracy of detection. Nowadays, deep learning methodologies have been used in medical field to diagnose the health conditions precisely and aid the clinicians. In this study, a new deep one-dimensional convolutional neural network (1D CNN) model is proposed for the automatic recognition of normal and abnormal EEG signals. The proposed model is a complete end-to-end structure which classifies the EEG signals without requiring any feature extraction. In this study, we have used the EEG signals from temporal to occipital (T5–O1) single channel obtained from Temple University Hospital EEG Abnormal Corpus (v2.0.0) EEG dataset to develop the 1D CNN model. Our developed model has yielded the classification error rate of 20.66% in classifying the normal and abnormal EEG signals.

  • 
  AtıfScopus164·OpenAlex211

  İncele
• MakaleYalnız künye2019Scopus SJRQ1/r/munzur/3595
  Deep long short-term memory networks-based automatic recognition of six different digital modulation types under varying noise conditions

  Daldal, Nihat; Yıldırım, Özal; Polat, Kemal

  In this paper, a new method based on deep learning has been proposed in order to recognize noise-digital modulation signals at varying noise levels automatically. The 8-bit data from six different modulations have been obtained by adding noise levels from 5 to 25 dB. The used digital modulation types are Amplitude Shift Keying, Frequency Shift Keying, Phase Shift Keying, Quadrature Amplitude Shift Keying, Quadrature Frequency Shift Keying, and Quadrature Phase Shift Keying. To recognize the noise-digital modulation signals automatically, a new deep long short-term memory networks (LSTMs) model has been proposed and then applied to these signals successfully. A significant advantage of the proposed system is that deep learning method has been trained and tested with raw digital modulation signals without applying any feature extraction from the signals. In this study, the noise modulation signals of 5–25 dB have been classified and compared with each other. The innovative aspect of the study is to classify the modulation with the LSTM method without dealing with the extraction of signal characteristics. Without noise, added digital modulation signals had been classified as the success rate of 97.22%, while with all noise-added signals have been classified as the success rate of 94.72% with deep LSTM model. The experimental results show that the proposed deep LSTM model has been achieved remarkable results in recognition of noised six different modulation signals with a fully end-to-end structure.

  AtıfScopus35·WoS26·OpenAlex35

  İncele