Mustafa Açıkkar
Research Subjects
Mustafa Açıkkar has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
Mehmet Fatih Akay
Bilgisayar Mühendisliği Bölümü, Çukurova Üniversitesi, Adana, Türkiye
3 shared publicationsErman Aktürk
Bilgisayar Mühendisliği Bölümü, Çukurova Üniversitesi, Adana, Türkiye
2 shared publications- Selçuk Tokgöz
Department of Software Engineering, Faculty of Computer and Informatics, Adana Alparslan Turkes Science and Technology University, Adana, Turkey
2 shared publications - Turgay Ibrikci
Çukurova University
2 shared publications - Baran Aydın
Adana Science and Technology University
1 shared publication Ece Aktarla
Matematik-Bilgisayar Bölümü, Çağ Üniversitesi, Mersin, TÜRKİYE
1 shared publication
Latest Publications
- 16 December 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 55 (3) , 1-25
Improving multi-class classification: scaled extensions of harmonic mean-based adaptive k-nearest neighbors
- 16 December 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 55 (3) , 1-25
Improving multi-class classification: scaled extensions of harmonic mean-based adaptive k-nearest neighbors
- Open Access
- 9 December 2024
- Published by Elsevier in Applied Ocean Research
- Vol. 154
Deep learning-based landslide tsunami run-up prediction from synthetic gage data
- 9 December 2024
- Published by Elsevier in Applied Ocean Research
- Vol. 154
Open AccessDeep learning-based landslide tsunami run-up prediction from synthetic gage data
- Open Access
- 7 February 2024
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 32 (1) , 68-92
Fast grid search: A grid search-inspired algorithm for optimizing hyperparameters of support vector regression
This study presents a fast hyperparameter optimization algorithm based on the benefits and shortcomings of the standard grid search (GS) algorithm for support vector regression (SVR). This presented GS-inspired algorithm, called fast grid search (FGS), was tested on benchmark datasets, and the impact of FGS on prediction accuracy was primarily compared with the GS algorithm on which it is based. To validate the efficacy of the proposed algorithm and conduct a comprehensive comparison, two additional hyperparameter optimization techniques, namely particle swarm optimization and Bayesian optimization, were also employed in the development of models on the given datasets. The evaluation of the models’ predictive performance was conducted by assessing root mean square error, mean absolute error, and mean absolute percentage error. In addition to these metrics, the number of evaluated submodels and the time required for optimization were used as determinative performance measures of the presented models. Experimental results proved that the FGS-optimized SVR models yield precise performance, supporting the reliability, validity, and applicability of the FGS algorithm. As a result, the FGS algorithm can be offered as a faster alternative in optimizing the hyperparameters of SVR in terms of execution time.
- 7 February 2024
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 32 (1) , 68-92
Open AccessFast grid search: A grid search-inspired algorithm for optimizing hyperparameters of support vector regression
This study presents a fast hyperparameter optimization algorithm based on the benefits and shortcomings of the standard grid search (GS) algorithm for support vector regression (SVR). This presented GS-inspired algorithm, called fast grid search (FGS), was tested on benchmark datasets, and the impact of FGS on prediction accuracy was primarily compared with the GS algorithm on which it is based. To validate the efficacy of the proposed algorithm and conduct a comprehensive comparison, two additional hyperparameter optimization techniques, namely particle swarm optimization and Bayesian optimization, were also employed in the development of models on the given datasets. The evaluation of the models’ predictive performance was conducted by assessing root mean square error, mean absolute error, and mean absolute percentage error. In addition to these metrics, the number of evaluated submodels and the time required for optimization were used as determinative performance measures of the presented models. Experimental results proved that the FGS-optimized SVR models yield precise performance, supporting the reliability, validity, and applicability of the FGS algorithm. As a result, the FGS algorithm can be offered as a faster alternative in optimizing the hyperparameters of SVR in terms of execution time.
- 8 December 2023
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (8) , 4027-4045
An improved KNN classifier based on a novel weighted voting function and adaptive k-value selection
- 8 December 2023
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (8) , 4027-4045
An improved KNN classifier based on a novel weighted voting function and adaptive k-value selection
- 14 July 2023
- Published by Springer Nature in Neural Computing and Applications
- Vol. 35 (27) , 19961-19977
A novel hybrid PSO- and GS-based hyperparameter optimization algorithm for support vector regression
- 14 July 2023
- Published by Springer Nature in Neural Computing and Applications
- Vol. 35 (27) , 19961-19977
A novel hybrid PSO- and GS-based hyperparameter optimization algorithm for support vector regression
- Open Access
- 22 November 2022
- Published by Springer Nature in Research Square
A Novel Hybrid PSO- and GS-based Hyperparameter Optimization Algorithm for Support Vector Regression
Hyperparameter optimization is vital in improving the prediction accuracy of Support Vector Regression (SVR), as in all machine learning algorithms. This study introduces a new hybrid optimization algorithm, namely PSOGS, which consolidates two strong and widely used algorithms, Particle Swarm Optimization (PSO) and Grid Search (GS). This hybrid algorithm was experimented on five benchmark datasets. The speed and the prediction accuracy of PSOGS-optimized SVR models (PSOGS-SVR) were compared to those of its constituent algorithms (PSO and GS) and another hybrid optimization algorithm (PSOGSA) that combines PSO and Gravitational Search Algorithm (GSA). The prediction accuracies were evaluated and compared in terms of root mean square error and mean absolute percentage error. For the sake of reliability, the results of the experiments were obtained by performing 10-fold cross-validation on 30 runs. The results showed that PSOGS-SVR yields prediction accuracy comparable to GS-SVR, performs much faster than GS-SVR, and provides better results with less execution time than PSO-SVR. Besides, PSOGS-SVR presents more effective results than PSOGSA-SVR in terms of both prediction accuracy and execution time. As a result, this study proved that PSOGS is a fast, stable, efficient, and reliable algorithm for optimizing hyperparameters of SVR.
- 22 November 2022
- Published by Springer Nature in Research Square
Open AccessA Novel Hybrid PSO- and GS-based Hyperparameter Optimization Algorithm for Support Vector Regression
Hyperparameter optimization is vital in improving the prediction accuracy of Support Vector Regression (SVR), as in all machine learning algorithms. This study introduces a new hybrid optimization algorithm, namely PSOGS, which consolidates two strong and widely used algorithms, Particle Swarm Optimization (PSO) and Grid Search (GS). This hybrid algorithm was experimented on five benchmark datasets. The speed and the prediction accuracy of PSOGS-optimized SVR models (PSOGS-SVR) were compared to those of its constituent algorithms (PSO and GS) and another hybrid optimization algorithm (PSOGSA) that combines PSO and Gravitational Search Algorithm (GSA). The prediction accuracies were evaluated and compared in terms of root mean square error and mean absolute percentage error. For the sake of reliability, the results of the experiments were obtained by performing 10-fold cross-validation on 30 runs. The results showed that PSOGS-SVR yields prediction accuracy comparable to GS-SVR, performs much faster than GS-SVR, and provides better results with less execution time than PSO-SVR. Besides, PSOGS-SVR presents more effective results than PSOGSA-SVR in terms of both prediction accuracy and execution time. As a result, this study proved that PSOGS is a fast, stable, efficient, and reliable algorithm for optimizing hyperparameters of SVR.
- Open Access
- 1 April 2020
- Published by ASOS Yayinevi in European Journal of Science and Technology
- No. 18,p. 16-24
Yıkanmış Türk Linyit Kömürlerinin Üst Isıl Değerinin Destek Vektör Regresyonu ile Tahmini
Bu çalışmada yıkanmış Türk linyit kömürlerinin üst ısıl değeri (GCV), makine öğrenmesi yöntemleri ile kömür numunelerinin kuru baz kısa analiz sonuçları kullanılarak tahmin edilmiştir. Laboratuvar kömür analiz sonuçlarından elde edilen kül (A), uçucu madde (VM), kükürt (S) ve GCV değişkenleri kullanılarak veri kümesi oluşturulmuştur. Veri kümesine, Destek Vektör Regresyonu (SVR) ile Çok Katmanlı Algılayıcı (MLP), Genel Regresyon Sinir Ağı (GRNN) ve Radyal Temelli Fonksiyon Sinir Ağı (RBFN) olmak üzere üç farklı Yapay Sinir Ağı (ANN) uygulanarak GCV tahmin modelleri geliştirilmiştir. Geliştirilen modellerin performans genelleştirme kabiliyeti 10-katlı çapraz-doğrulama kullanılarak sağlanmış ve modellerin tahmin doğruluğu, performans ölçütleri Çoklu Korelasyon Katsayısı (R), Kök Ortalama Kare Hatası (RMSE), Ortalama Mutlak Hata (MAE) ve Ortalama Mutlak Yüzde Hata (MAPE) kullanılarak hesaplanmıştır. Sonuçlar, GCV tahmini için, tüm modeller arasında SVR tabanlı modelin ANN tabanlı modellere göre biraz daha iyi, ANN tabanlı modeller arasında ise RBFN tabanlı modelin MLP ve GRNN tabanlı modellere göre daha iyi performans gösterdiğini ortaya koymuştur.
- 1 April 2020
- Published by ASOS Yayinevi in European Journal of Science and Technology
- No. 18,p. 16-24
Open AccessYıkanmış Türk Linyit Kömürlerinin Üst Isıl Değerinin Destek Vektör Regresyonu ile Tahmini
Bu çalışmada yıkanmış Türk linyit kömürlerinin üst ısıl değeri (GCV), makine öğrenmesi yöntemleri ile kömür numunelerinin kuru baz kısa analiz sonuçları kullanılarak tahmin edilmiştir. Laboratuvar kömür analiz sonuçlarından elde edilen kül (A), uçucu madde (VM), kükürt (S) ve GCV değişkenleri kullanılarak veri kümesi oluşturulmuştur. Veri kümesine, Destek Vektör Regresyonu (SVR) ile Çok Katmanlı Algılayıcı (MLP), Genel Regresyon Sinir Ağı (GRNN) ve Radyal Temelli Fonksiyon Sinir Ağı (RBFN) olmak üzere üç farklı Yapay Sinir Ağı (ANN) uygulanarak GCV tahmin modelleri geliştirilmiştir. Geliştirilen modellerin performans genelleştirme kabiliyeti 10-katlı çapraz-doğrulama kullanılarak sağlanmış ve modellerin tahmin doğruluğu, performans ölçütleri Çoklu Korelasyon Katsayısı (R), Kök Ortalama Kare Hatası (RMSE), Ortalama Mutlak Hata (MAE) ve Ortalama Mutlak Yüzde Hata (MAPE) kullanılarak hesaplanmıştır. Sonuçlar, GCV tahmini için, tüm modeller arasında SVR tabanlı modelin ANN tabanlı modellere göre biraz daha iyi, ANN tabanlı modeller arasında ise RBFN tabanlı modelin MLP ve GRNN tabanlı modellere göre daha iyi performans gösterdiğini ortaya koymuştur.
- Open Access
- 28 September 2018
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 26 (5) , 2541-2552
Prediction of gross calorific value of coal based on proximate analysis using multiple linear regression and artificial neural networks
- 28 September 2018
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 26 (5) , 2541-2552
Open AccessPrediction of gross calorific value of coal based on proximate analysis using multiple linear regression and artificial neural networks
- 1 April 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Intelligent regression techniques for non-exercise prediction of VO2max
The purpose of this study is to develop non-exercise (N-Ex) VO 2 max prediction models by using Support Vector Regression (SVR) and Multilayer Feed Forward Neural Networks (MFFNN). VO 2 max values of 100 subjects are measured using a maximal graded exercise test. The variables; gender, age, body mass index (BMI), perceived functional ability (PFA) to walk, jog or run given distances and current physical activity rating (PA-R) are used to build two N-Ex prediction models. Using 10-fold cross validation on the dataset, standard error of estimates (SEE) and multiple correlation coefficients (R) of both models are calculated. The MFFNN-based model yields lower SEE (3.23 ml·kg -1 ·min -1 ) whereas the SVR-based model yields higher R (0.93). Compared with the results of the other N-Ex prediction models in literature that are developed using Multiple Linear Regression Analysis, the reported values of SEE and R in this study are considerably more accurate.
- 1 April 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Intelligent regression techniques for non-exercise prediction of VO2max
The purpose of this study is to develop non-exercise (N-Ex) VO 2 max prediction models by using Support Vector Regression (SVR) and Multilayer Feed Forward Neural Networks (MFFNN). VO 2 max values of 100 subjects are measured using a maximal graded exercise test. The variables; gender, age, body mass index (BMI), perceived functional ability (PFA) to walk, jog or run given distances and current physical activity rating (PA-R) are used to build two N-Ex prediction models. Using 10-fold cross validation on the dataset, standard error of estimates (SEE) and multiple correlation coefficients (R) of both models are calculated. The MFFNN-based model yields lower SEE (3.23 ml·kg -1 ·min -1 ) whereas the SVR-based model yields higher R (0.93). Compared with the results of the other N-Ex prediction models in literature that are developed using Multiple Linear Regression Analysis, the reported values of SEE and R in this study are considerably more accurate.
- 1 April 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Neural network based VO2max prediction models using maximal exercise and non-exercise data
Artificial Neural Network (ANN) models based on maximal and non-exercise (N-Ex) variables are developed to predict maximal oxygen uptake (VO2max) the input variables of the dataset are gender, age, body mass index (BMI), grade, self-reported rating of perceived exertion (RPE) from treadmill test, heart rate (HR), perceived functional ability (PFA) and physical activity rating (PA-R). The performance of the models is evaluated by calculating their standard error of estimate (SEE) and multiple correlation coefficient (R). The results suggest that the performance of VO 2 max prediction models based on maximal and standard N-Ex variables (i.e. gender, age, BMI etc) can be improved by including questionnaire variables (PFA and PA-R) in the models.
- 1 April 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Neural network based VO2max prediction models using maximal exercise and non-exercise data
Artificial Neural Network (ANN) models based on maximal and non-exercise (N-Ex) variables are developed to predict maximal oxygen uptake (VO2max) the input variables of the dataset are gender, age, body mass index (BMI), grade, self-reported rating of perceived exertion (RPE) from treadmill test, heart rate (HR), perceived functional ability (PFA) and physical activity rating (PA-R). The performance of the models is evaluated by calculating their standard error of estimate (SEE) and multiple correlation coefficient (R). The results suggest that the performance of VO 2 max prediction models based on maximal and standard N-Ex variables (i.e. gender, age, BMI etc) can be improved by including questionnaire variables (PFA and PA-R) in the models.