İlker Ercanlı
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
İlker Ercanlı has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Alkan Günlü
Çankırı Karatekin University
4 shared publications - Ferhat Bolat
Cankiri Karatekin Universitesi
4 shared publications - Emin Zeki Baskent
Faculty of Forestry, Karadeniz Technical University, 61080 Trabzon, Turkey
3 shared publications Ferhat Bolat
3 shared publications- Muammer Senyurt
AÇankırı Karatekin University, Faculty of Forestry, Department of Forest Engineering, Çankırı, Turkey, phone +90 3762122757
3 shared publications - Abdurrahman Şahi̇n
Artvin Çoruh Üniversitesi
2 shared publications
Latest Publications
- 29 December 2025
- Published by Turkish Journal of Forestry in Turkish Journal of Forestry | Türkiye Ormancılık Dergisi
- Vol. 26 (4) , 460-474
Sisorta Orman İşletme Şefliği sınırları içerisinde yer alan sarıçam ağaçlarının hacimlerinin tahmininde Çok Değişkenli Uyarlanabilir Regresyon Eğrileri Tekniğinin kullanımı
Bu çalışma, Giresun Orman Bölge Müdürlüğü, Koyulhisar Orman İşletme Müdürlüğü, Sisorta Orman İşletme Şefliği sınırları içerisinde yer alan sarıçam (Pinus sylvestris L.) meşcerelerinde tek ağaç gövde hacimlerinin elde edilmesinde, tek ve çift girişli ağaç hacim tahmin modellerinin geliştirilmesi ile Çok Değişkenli Uyarlanabilir Regresyon Eğrileri (Multivariate Adaptive Regression Splines – MARS) yöntemine dayana tahmin modellerinin geliştirilmesi amaçlanmıştır. Bu amaçla, farklı göğüs çapları ve boylarda olmak üzere ağaçların gövde hacmindeki değişimi temsil etmek üzere alınmış 238 örnek ağaçtan elde edilen veriler kullanılmış ve bu veriler, tek ve çift girişli ağaç hacim denklemleri ile MARS modellerinin geliştirilmesi (n=190) ve geliştirilen bu modellerin çalışma alanındaki meşcerelere uygunluğunun test edilmesi (n=48) amacıyla rastgele örnekleme yöntemiyle iki alt gruba ayrılmıştır. Jerome Friedman tarafından 1991 yılında geliştirilen parametrik olmayan bir veri madenciliği yöntemi olan MARS tekniği, herhangi bir istatistiksel varsayıma ihtiyaç duymadan karmaşık doğrusal olmayan ilişkileri başarılı bir şekilde modelleyebilmektedir. Tek girişli tahminlerde, çoklu regresyon denklemi ile THY=%0.07260, OMHY=%10.7788 ve R2=0.9696 olarak sonuçlar elde edilirken; tek girişli MARS modeli THY'yi % 9.74x10-7, OMHY'yi %10.6510 ve R2= 0.9736 olarak hesaplamıştır. Çift girişli tahminlerde ise, çoklu regresyon denklemi ile THY=%0.31654, OMHY=%6.7784 ve R2= 0.9865 olarak elde edilmiş iken, çift girişli MARS modeli ile THY=%0.00099, OMHY=%5.8994 ve R2= 0.9885 olarak elde edilmiştir. Ayrıca, modellerin geliştirme sürecinde kullanılmayan bağımsız örnek ağaçlar kullanılarak yapılan denetlemede, Sisorta yöresi sarıçam meşcereleri için mevcut amenajman hacim tablosunun ve geliştirilen tek girişli denklemin istatistiksel olarak %95 güvenle uygun olmadığı; çoklu regresyon analizi ile elde edilen çift girişli hacim denkleminin ve hem tek hem de çift girişli MARS modellerinin uygun olduğu belirlenmiştir. Bu bulgularla, MARS tekniğinin parçalı doğrusal segmentler ve düğüm noktaları aracılığıyla değişkenler arası karmaşık ilişkileri esnek bir şekilde modelleme yeteneği, ormancılık uygulamalarında önemli bir potansiyel sunduğu sonucuna varılabilir. Bu yenilikçi yaklaşımın Türkiye'nin farklı ormanlık alanlarında yaygınlaştırılması, orman amenajmanı ve envanter çalışmalarının başarısına önemli katkılar sağlayabilecektir.
- 29 December 2025
- Published by Turkish Journal of Forestry in Turkish Journal of Forestry | Türkiye Ormancılık Dergisi
- Vol. 26 (4) , 460-474
Sisorta Orman İşletme Şefliği sınırları içerisinde yer alan sarıçam ağaçlarının hacimlerinin tahmininde Çok Değişkenli Uyarlanabilir Regresyon Eğrileri Tekniğinin kullanımı
Bu çalışma, Giresun Orman Bölge Müdürlüğü, Koyulhisar Orman İşletme Müdürlüğü, Sisorta Orman İşletme Şefliği sınırları içerisinde yer alan sarıçam (Pinus sylvestris L.) meşcerelerinde tek ağaç gövde hacimlerinin elde edilmesinde, tek ve çift girişli ağaç hacim tahmin modellerinin geliştirilmesi ile Çok Değişkenli Uyarlanabilir Regresyon Eğrileri (Multivariate Adaptive Regression Splines – MARS) yöntemine dayana tahmin modellerinin geliştirilmesi amaçlanmıştır. Bu amaçla, farklı göğüs çapları ve boylarda olmak üzere ağaçların gövde hacmindeki değişimi temsil etmek üzere alınmış 238 örnek ağaçtan elde edilen veriler kullanılmış ve bu veriler, tek ve çift girişli ağaç hacim denklemleri ile MARS modellerinin geliştirilmesi (n=190) ve geliştirilen bu modellerin çalışma alanındaki meşcerelere uygunluğunun test edilmesi (n=48) amacıyla rastgele örnekleme yöntemiyle iki alt gruba ayrılmıştır. Jerome Friedman tarafından 1991 yılında geliştirilen parametrik olmayan bir veri madenciliği yöntemi olan MARS tekniği, herhangi bir istatistiksel varsayıma ihtiyaç duymadan karmaşık doğrusal olmayan ilişkileri başarılı bir şekilde modelleyebilmektedir. Tek girişli tahminlerde, çoklu regresyon denklemi ile THY=%0.07260, OMHY=%10.7788 ve R2=0.9696 olarak sonuçlar elde edilirken; tek girişli MARS modeli THY'yi % 9.74x10-7, OMHY'yi %10.6510 ve R2= 0.9736 olarak hesaplamıştır. Çift girişli tahminlerde ise, çoklu regresyon denklemi ile THY=%0.31654, OMHY=%6.7784 ve R2= 0.9865 olarak elde edilmiş iken, çift girişli MARS modeli ile THY=%0.00099, OMHY=%5.8994 ve R2= 0.9885 olarak elde edilmiştir. Ayrıca, modellerin geliştirme sürecinde kullanılmayan bağımsız örnek ağaçlar kullanılarak yapılan denetlemede, Sisorta yöresi sarıçam meşcereleri için mevcut amenajman hacim tablosunun ve geliştirilen tek girişli denklemin istatistiksel olarak %95 güvenle uygun olmadığı; çoklu regresyon analizi ile elde edilen çift girişli hacim denkleminin ve hem tek hem de çift girişli MARS modellerinin uygun olduğu belirlenmiştir. Bu bulgularla, MARS tekniğinin parçalı doğrusal segmentler ve düğüm noktaları aracılığıyla değişkenler arası karmaşık ilişkileri esnek bir şekilde modelleme yeteneği, ormancılık uygulamalarında önemli bir potansiyel sunduğu sonucuna varılabilir. Bu yenilikçi yaklaşımın Türkiye'nin farklı ormanlık alanlarında yaygınlaştırılması, orman amenajmanı ve envanter çalışmalarının başarısına önemli katkılar sağlayabilecektir.
- 1 December 2024
- Published by Canadian Science Publishing in Canadian Journal of Forest Research
- Vol. 54 (12) , 1500-1518
Deep learning algorithms for addressing overfitting and biological realism in tree taper and volume predictions
This study addresses the challenges of overfitting and maintaining biological realism in deep learning algorithms (DLAs), for predicting individual tree taper using stem diameters outside bark (DOB) and total tree volume (TTV). To this end, DLAs were trained using two different approaches: a “hyperparameter-optimized DLA”, which customizes specific hyperparameters such as learning rate and momentum rate, and a “regularization-optimized DLA”, which incorporates optimization techniques like early stopping with root mean square error, L1 and L2 regularization, and dropout. Although obtaining the deterioration in predictive capabilities statistics from the taring dataset to the validation dataset by standard DLA with adaptive learning processes without customizing the hyperparameters and regularization parameters, the hyperparameter-optimized DLA with a momentum of 0.8, and a 7 # hidden layer for the TTV and regularization-optimized DLA with a dropout ratio of 0.000001, a 3 # hidden layer for the DOB demonstrated comparable predictive capabilities statistics across both training and validation datasets with generating biologically plausible predictions. Our results support that these hyperparameter-optimized and regularization-optimized DLAs, by improving the “black-box” nature of artificial intelligence, offer significant potential for enhanced interpretability and performance by improving the problem of overfitting and the violations biological realism in forest biometrics applications.
- 1 December 2024
- Published by Canadian Science Publishing in Canadian Journal of Forest Research
- Vol. 54 (12) , 1500-1518
Deep learning algorithms for addressing overfitting and biological realism in tree taper and volume predictions
This study addresses the challenges of overfitting and maintaining biological realism in deep learning algorithms (DLAs), for predicting individual tree taper using stem diameters outside bark (DOB) and total tree volume (TTV). To this end, DLAs were trained using two different approaches: a “hyperparameter-optimized DLA”, which customizes specific hyperparameters such as learning rate and momentum rate, and a “regularization-optimized DLA”, which incorporates optimization techniques like early stopping with root mean square error, L1 and L2 regularization, and dropout. Although obtaining the deterioration in predictive capabilities statistics from the taring dataset to the validation dataset by standard DLA with adaptive learning processes without customizing the hyperparameters and regularization parameters, the hyperparameter-optimized DLA with a momentum of 0.8, and a 7 # hidden layer for the TTV and regularization-optimized DLA with a dropout ratio of 0.000001, a 3 # hidden layer for the DOB demonstrated comparable predictive capabilities statistics across both training and validation datasets with generating biologically plausible predictions. Our results support that these hyperparameter-optimized and regularization-optimized DLAs, by improving the “black-box” nature of artificial intelligence, offer significant potential for enhanced interpretability and performance by improving the problem of overfitting and the violations biological realism in forest biometrics applications.
- Open Access
- 1 January 2024
- Published by Cankiri Karatekin Universitesi in Anadolu Orman Araştırmaları Dergisi
- Vol. 9 (2) , 126-134
ÇAKÜ Araştırma Ormanı Karaçam meşcereleri için tek ve çift girişli toprak üstü ağaç biyokütle denklemlerinin geliştirilmesi
Bu çalışmada, Çankırı Orman İşletme Müdürlüğü sınırları içerisinde yer alan Çankırı Karatekin Üniversitesi Orman Fakültesi Araştırma ve Uygulama Ormanı içerisinde bulunan saf ve aynı yaşlı Anadolu Karaçamı meşcerelerinden seçilen 50 adet örnek ağaç kullanılarak tek ve çift girişli gövde, kabuk, ibre, dal ve toplam toprak üstü biyokütle denklemleri geliştirilmiştir. Geliştirilen bu denklemlerden, tek girişli gövde biyokütle denkleminin belirtme katsayısı (R2), toplam hata yüzdesi (THY), ortalama mutlak hata yüzdesi (OMHY) değerleri sırasıyla 0,911, -4,7763 ve 16,9823 olarak hesaplanmıştır. Bu değerler çift girişli gövde biyokütle denklemi için sırasıyla 0,951, 0,0000 ve 12,2506 olarak bulunmuştur. Tek ve çift girişli kabuk biyokütle denklemlerinin R2, THY ve OMHY değerleri (devam eden cümlelerde köşeli parantez içerisinde verilen değerler çift girişli denklemlere ait, []) sırasıyla 0,885 [0,926], -4,2410 [0,0000] ve 20,3853 [17,6638]’tür. Tek ve çift girişli ibre biyokütle denklemi için benzer şekilde R2, THY ve OMHY değerleri sırasıyla 0,547 [0,637], ve 31,4892 [0,0000] ve 55,3247 [57,1074] olarak elde edilmiştir. Tek ve çift girişli dal biyokütle denklemlerinin R2, THY ve OMHY değerleri sırasıyla 0,781 [0,793], 0,5932 [6,2919] ve 33,5256 [34,7849] olarak hesaplanmıştır. Tek ve çift girişli toplam toprak üstü biyokütle denklemlerinin R2, THY ve OMHY değerleri ise sırasıyla 0,913 [0,954], 2,3304 [7,5169] ve 17,5011 [14,5473] olarak elde edilmiştir. Tüm bu denklemlere ilişkin bütün parametreler, istatistiksel olarak p
- 1 January 2024
- Published by Cankiri Karatekin Universitesi in Anadolu Orman Araştırmaları Dergisi
- Vol. 9 (2) , 126-134
Open AccessÇAKÜ Araştırma Ormanı Karaçam meşcereleri için tek ve çift girişli toprak üstü ağaç biyokütle denklemlerinin geliştirilmesi
Bu çalışmada, Çankırı Orman İşletme Müdürlüğü sınırları içerisinde yer alan Çankırı Karatekin Üniversitesi Orman Fakültesi Araştırma ve Uygulama Ormanı içerisinde bulunan saf ve aynı yaşlı Anadolu Karaçamı meşcerelerinden seçilen 50 adet örnek ağaç kullanılarak tek ve çift girişli gövde, kabuk, ibre, dal ve toplam toprak üstü biyokütle denklemleri geliştirilmiştir. Geliştirilen bu denklemlerden, tek girişli gövde biyokütle denkleminin belirtme katsayısı (R2), toplam hata yüzdesi (THY), ortalama mutlak hata yüzdesi (OMHY) değerleri sırasıyla 0,911, -4,7763 ve 16,9823 olarak hesaplanmıştır. Bu değerler çift girişli gövde biyokütle denklemi için sırasıyla 0,951, 0,0000 ve 12,2506 olarak bulunmuştur. Tek ve çift girişli kabuk biyokütle denklemlerinin R2, THY ve OMHY değerleri (devam eden cümlelerde köşeli parantez içerisinde verilen değerler çift girişli denklemlere ait, []) sırasıyla 0,885 [0,926], -4,2410 [0,0000] ve 20,3853 [17,6638]’tür. Tek ve çift girişli ibre biyokütle denklemi için benzer şekilde R2, THY ve OMHY değerleri sırasıyla 0,547 [0,637], ve 31,4892 [0,0000] ve 55,3247 [57,1074] olarak elde edilmiştir. Tek ve çift girişli dal biyokütle denklemlerinin R2, THY ve OMHY değerleri sırasıyla 0,781 [0,793], 0,5932 [6,2919] ve 33,5256 [34,7849] olarak hesaplanmıştır. Tek ve çift girişli toplam toprak üstü biyokütle denklemlerinin R2, THY ve OMHY değerleri ise sırasıyla 0,913 [0,954], 2,3304 [7,5169] ve 17,5011 [14,5473] olarak elde edilmiştir. Tüm bu denklemlere ilişkin bütün parametreler, istatistiksel olarak p
- Open Access
- 28 August 2023
- Published by Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria (INIA) in Forest Systems
- Vol. 32 (3) , e016
An evaluation of various probability density functions for predicting diameter distributions in pure and mixed-species stands in Türkiye
Aim of study: To assess the capabilities of some infrequently used probability density functions (PDFs) in modeling stand diameter distributions and compare their performance to that of typical PDFs. Area of study: The research was conducted in pure and mixed stands located in the OF Planning Unit of the Trabzon Forest Regional Directorate in Northern Türkiye. Material and methods: A set of 17,324 DBH measurements, originating from 608 sample plots located in stands of even-aged and pure and mixed stands, were obtained to represent various stand conditions such as site quality, age, and stand density in OF planning unit forests. In order to ensure a minimum of 30-40 trees in each sample plot, the plot sizes ranged from 0.04 to 0.08 hectares, depending on stand density. The parameters of PDFs include Weibull with 3P and 2P, Rice, Rayleigh, Normal, Nagakami, Lognormal with 2P and 3P, Lévy with 1p and 2P, Laplace, Kumaraswamy, Johnson’s SB, and Gamma were estimated using the maximum likelihood estimation (MLE) prediction procedure. Additionally, the goodness of fit test was combined with the Kolmogorov-Smirnov test (statistically at a 95% confidence interval). Main results: The Rayleigh distribution was the model that best explained the diameter distributions of pure and mixed forests in the OF Planning Unit (as Fit Index (FI) = 0.6743 and acceptance rate 96.4% based on the result of one sample Kolmogorov-Smirnov test). Research highlights: Less commonly used PDFs such as Rice, Nakagami, and Kumaraswamy-4P demonstrated superior predictive performance compared to some traditional distributions widely used in forestry, including Weibull-2P and -3P, Johnson’s SB, Normal, Gamma-3P, and Lognormal-3P.
- 28 August 2023
- Published by Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria (INIA) in Forest Systems
- Vol. 32 (3) , e016
Open AccessAn evaluation of various probability density functions for predicting diameter distributions in pure and mixed-species stands in Türkiye
Aim of study: To assess the capabilities of some infrequently used probability density functions (PDFs) in modeling stand diameter distributions and compare their performance to that of typical PDFs. Area of study: The research was conducted in pure and mixed stands located in the OF Planning Unit of the Trabzon Forest Regional Directorate in Northern Türkiye. Material and methods: A set of 17,324 DBH measurements, originating from 608 sample plots located in stands of even-aged and pure and mixed stands, were obtained to represent various stand conditions such as site quality, age, and stand density in OF planning unit forests. In order to ensure a minimum of 30-40 trees in each sample plot, the plot sizes ranged from 0.04 to 0.08 hectares, depending on stand density. The parameters of PDFs include Weibull with 3P and 2P, Rice, Rayleigh, Normal, Nagakami, Lognormal with 2P and 3P, Lévy with 1p and 2P, Laplace, Kumaraswamy, Johnson’s SB, and Gamma were estimated using the maximum likelihood estimation (MLE) prediction procedure. Additionally, the goodness of fit test was combined with the Kolmogorov-Smirnov test (statistically at a 95% confidence interval). Main results: The Rayleigh distribution was the model that best explained the diameter distributions of pure and mixed forests in the OF Planning Unit (as Fit Index (FI) = 0.6743 and acceptance rate 96.4% based on the result of one sample Kolmogorov-Smirnov test). Research highlights: Less commonly used PDFs such as Rice, Nakagami, and Kumaraswamy-4P demonstrated superior predictive performance compared to some traditional distributions widely used in forestry, including Weibull-2P and -3P, Johnson’s SB, Normal, Gamma-3P, and Lognormal-3P.
- Open Access
- 30 June 2023
- Published by Ormancilik Arastirma Dergisi in Ormancılık Araştırma Dergisi
- Vol. 10 (1) , 32-44
Karışık etkili modelleme yaklaşımıyla karaçam hacim denklemlerinin geliştirilmesi
Bu çalışmada, İç Anadolu ve Akdeniz Bölgelerindeki dört farklı yörede bulunan saf Anadolu Karaçamı meşcereleri için bölgesel tek ve çift girişli kabuklu ağaç hacim denklemleri geliştirilmiştir. Bu amaçla, ölçülen ağaçlardan 170’i denklem geliştirmede, 30’u ise kontrolde kullanılmıştır. Denklem geliştirmek için 4 adet tek ve 16 adet çift girişli ağaç hacim denklemi, 6 farklı başarı ölçütüne göre kıyaslanmış ve en başarılı olan denklemler belirlenmiştir. Seçilen denklemlerin düzeltilmiş belirtme katsayısı (R2düz), hata kareler ortalamasının karekökü (HKOK), ortalama mutlak hata yüzdesi (%OMH) ve toplam hata yüzdesi (%TH) değerleri sırasıyla; tek girişli denklemde 0,951; 0,145; 12,798 ve 0,263 şeklinde, çift girişli denklemde ise 0,976; 0,103; 8,978 ve 6x10-4 olarak hesaplanmıştır. Ayrıca karışık etkili modelleme yaklaşımıyla, elde edilen hacim denklemlerinin katsayıları, her yöre için ayrı ayrı yeniden hesaplanmış ve hacim gelişimindeki bölgesel farklılık ortaya konmuştur. Son olarak da, geliştirilen denklemlerin daha önceki karaçam hacim denklemleriyle kıyaslaması yapılmıştır.
- 30 June 2023
- Published by Ormancilik Arastirma Dergisi in Ormancılık Araştırma Dergisi
- Vol. 10 (1) , 32-44
Open AccessKarışık etkili modelleme yaklaşımıyla karaçam hacim denklemlerinin geliştirilmesi
Bu çalışmada, İç Anadolu ve Akdeniz Bölgelerindeki dört farklı yörede bulunan saf Anadolu Karaçamı meşcereleri için bölgesel tek ve çift girişli kabuklu ağaç hacim denklemleri geliştirilmiştir. Bu amaçla, ölçülen ağaçlardan 170’i denklem geliştirmede, 30’u ise kontrolde kullanılmıştır. Denklem geliştirmek için 4 adet tek ve 16 adet çift girişli ağaç hacim denklemi, 6 farklı başarı ölçütüne göre kıyaslanmış ve en başarılı olan denklemler belirlenmiştir. Seçilen denklemlerin düzeltilmiş belirtme katsayısı (R2düz), hata kareler ortalamasının karekökü (HKOK), ortalama mutlak hata yüzdesi (%OMH) ve toplam hata yüzdesi (%TH) değerleri sırasıyla; tek girişli denklemde 0,951; 0,145; 12,798 ve 0,263 şeklinde, çift girişli denklemde ise 0,976; 0,103; 8,978 ve 6x10-4 olarak hesaplanmıştır. Ayrıca karışık etkili modelleme yaklaşımıyla, elde edilen hacim denklemlerinin katsayıları, her yöre için ayrı ayrı yeniden hesaplanmış ve hacim gelişimindeki bölgesel farklılık ortaya konmuştur. Son olarak da, geliştirilen denklemlerin daha önceki karaçam hacim denklemleriyle kıyaslaması yapılmıştır.
- Open Access
- 20 February 2023
- Published by Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria (INIA) in Forest Systems
- Vol. 32 (1) , e004
A comparison of artificial neural networks and regression modeling techniques for predicting dominant heights of Oriental spruce in a mixed stand
Aim of study: This paper introduces comparative evaluations of artificial neural network models and regression modeling techniques based on some fitting statistics and desirable characteristics for predicting dominant height. Area of study: The data of this study were obtained from Oriental spruce (Picea orientalis L.) felled trees in even-aged and mixed Oriental spruce and Scotch pine (Pinus sylvestris L.) stands in the northeast of Türkiye. Material and methods: A total of 873 height-age pairs were obtained from Oriental spruce trees in a mixed forest stand. Nonlinear mixed-effects models (NLMEs), autoregressive models (ARM), dummy variable method (DVM), and artificial neural networks (ANNs) were compared to predict dominant height growth. Main results: The best predictive model was NLME with a single random parameter (root mean square error, RMSE: 0.68 m). The results showed that NLMEs outperformed ARM (RMSE: 1.09 m), DVM in conjunction with ARM (RMSE: 1.09 m), and ANNs (RMSE: from 1.11 to 2.40 m) in the majority of the cases. Whereas considering variations among observations by random parameter(s) significantly improved predictions of dominant height, considering correlated error terms by autoregressive correlation parameter(s) enhanced slightly the predictions. ANNs generally underperformed compared to NLMEs, ARM, and DVM with ARM. Research highlights: All regression techniques fulfilled the desirable characteristics such as sigmoidal pattern, polymorphism, multiple asymptotes, base-age invariance, and inflection point. However, ANNs could not replicate most of these features, excluding the sigmoidal pattern. Accordingly, ANNs seem insufficient to assure biological growth assumptions regarding dominant height growth.
- 20 February 2023
- Published by Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria (INIA) in Forest Systems
- Vol. 32 (1) , e004
Open AccessA comparison of artificial neural networks and regression modeling techniques for predicting dominant heights of Oriental spruce in a mixed stand
Aim of study: This paper introduces comparative evaluations of artificial neural network models and regression modeling techniques based on some fitting statistics and desirable characteristics for predicting dominant height. Area of study: The data of this study were obtained from Oriental spruce (Picea orientalis L.) felled trees in even-aged and mixed Oriental spruce and Scotch pine (Pinus sylvestris L.) stands in the northeast of Türkiye. Material and methods: A total of 873 height-age pairs were obtained from Oriental spruce trees in a mixed forest stand. Nonlinear mixed-effects models (NLMEs), autoregressive models (ARM), dummy variable method (DVM), and artificial neural networks (ANNs) were compared to predict dominant height growth. Main results: The best predictive model was NLME with a single random parameter (root mean square error, RMSE: 0.68 m). The results showed that NLMEs outperformed ARM (RMSE: 1.09 m), DVM in conjunction with ARM (RMSE: 1.09 m), and ANNs (RMSE: from 1.11 to 2.40 m) in the majority of the cases. Whereas considering variations among observations by random parameter(s) significantly improved predictions of dominant height, considering correlated error terms by autoregressive correlation parameter(s) enhanced slightly the predictions. ANNs generally underperformed compared to NLMEs, ARM, and DVM with ARM. Research highlights: All regression techniques fulfilled the desirable characteristics such as sigmoidal pattern, polymorphism, multiple asymptotes, base-age invariance, and inflection point. However, ANNs could not replicate most of these features, excluding the sigmoidal pattern. Accordingly, ANNs seem insufficient to assure biological growth assumptions regarding dominant height growth.
- Open Access
- 23 November 2021
- Published by Springer Nature in Research Square
A Comprehensive Comparison of Artificial Neural Networks and Some Regression Modeling Techniques on Dominant Height Predictions of Oriental Spruce in a Mixed Forest, the Northeast Turkey
The authors have requested that this preprint be removed from Research Square.
- 23 November 2021
- Published by Springer Nature in Research Square
Open AccessA Comprehensive Comparison of Artificial Neural Networks and Some Regression Modeling Techniques on Dominant Height Predictions of Oriental Spruce in a Mixed Forest, the Northeast Turkey
The authors have requested that this preprint be removed from Research Square.
- 16 November 2020
- Published by Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria (INIA) in Forest Systems
- Vol. 29 (2) , e013
Artificial intelligence with deep learning algorithms to model relationships between total tree height and diameter at breast height
Aim of Study: As an innovative prediction technique, Artificial Intelligence technique based on a Deep Learning Algorithm (DLA) with various numbers of neurons and hidden layer alternatives were trained and evaluated to predict the relationships between total tree height (TTH) and diameter at breast height (DBH) with nonlinear least squared (NLS) regression models and nonlinear mixed effect (NLME) regression models.Area of Study: The data of this study were measured from even-aged, pure Turkish Pine (Pinus brutia Ten.) stands in the Kestel Forests located in the Bursa region of northwestern Turkey.Material and Methods: 1132 pairs of TTH-DBH measurements from 132 sample plots were used for modeling relationships between TTH, DBH, and stand attributes such as dominant height (Ho) and diameter (Do).Main Results: The combination of 100 # neurons and 8 # hidden layer in DLA resulted in the best predictive total height prediction values with Average Absolute Error (0.4188), max. Average Absolute Error (3.7598), Root Mean Squared Error (0.6942), Root Mean Squared error % (5.2164), Akaike Information Criteria (-345.4465), Bayesian Information Criterion (-330.836), the average Bias (0.0288) and the average Bias % (0.2166), and fitting abilities with r (0.9842) and Fit Index (0.9684). Also, the results of equivalence tests showed that the DLA technique successfully predicted the TTH in the validation dataset.Research highlights: These superior fitting scores coupled with the validation results in TTH predictions suggested that deep learning network models should be considered an alternative to the traditional nonlinear regression techniques and should be given importance as an innovative prediction technique.Keywords: Prediction; artificial intelligence; deep learning algorithms; number of neurons; hidden layer alternatives.Abbreviations: TTH (total tree height), DBH (diameter at breast height), OLS (ordinary least squares), NLME (nonlinear mixed effect), AIT (Artificial Intelligence Techniques), ANN (Artificial Neural Network), DLA (Deep Learning Algorithm), GPU (Graphical Processing Units), NLS (nonlinear least squared), RMSE (root mean squared error), AIC (Akaike information criteria), BIC (Bayesian information criterion), FI (fit index), AAE (average absolute error), BLUP (best linear unbiased predictor), TOST (two one-sided test method).
- 16 November 2020
- Published by Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria (INIA) in Forest Systems
- Vol. 29 (2) , e013
Artificial intelligence with deep learning algorithms to model relationships between total tree height and diameter at breast height
Aim of Study: As an innovative prediction technique, Artificial Intelligence technique based on a Deep Learning Algorithm (DLA) with various numbers of neurons and hidden layer alternatives were trained and evaluated to predict the relationships between total tree height (TTH) and diameter at breast height (DBH) with nonlinear least squared (NLS) regression models and nonlinear mixed effect (NLME) regression models.Area of Study: The data of this study were measured from even-aged, pure Turkish Pine (Pinus brutia Ten.) stands in the Kestel Forests located in the Bursa region of northwestern Turkey.Material and Methods: 1132 pairs of TTH-DBH measurements from 132 sample plots were used for modeling relationships between TTH, DBH, and stand attributes such as dominant height (Ho) and diameter (Do).Main Results: The combination of 100 # neurons and 8 # hidden layer in DLA resulted in the best predictive total height prediction values with Average Absolute Error (0.4188), max. Average Absolute Error (3.7598), Root Mean Squared Error (0.6942), Root Mean Squared error % (5.2164), Akaike Information Criteria (-345.4465), Bayesian Information Criterion (-330.836), the average Bias (0.0288) and the average Bias % (0.2166), and fitting abilities with r (0.9842) and Fit Index (0.9684). Also, the results of equivalence tests showed that the DLA technique successfully predicted the TTH in the validation dataset.Research highlights: These superior fitting scores coupled with the validation results in TTH predictions suggested that deep learning network models should be considered an alternative to the traditional nonlinear regression techniques and should be given importance as an innovative prediction technique.Keywords: Prediction; artificial intelligence; deep learning algorithms; number of neurons; hidden layer alternatives.Abbreviations: TTH (total tree height), DBH (diameter at breast height), OLS (ordinary least squares), NLME (nonlinear mixed effect), AIT (Artificial Intelligence Techniques), ANN (Artificial Neural Network), DLA (Deep Learning Algorithm), GPU (Graphical Processing Units), NLS (nonlinear least squared), RMSE (root mean squared error), AIC (Akaike information criteria), BIC (Bayesian information criterion), FI (fit index), AAE (average absolute error), BLUP (best linear unbiased predictor), TOST (two one-sided test method).
- Open Access
- 3 April 2020
- Published by Elsevier in Forest Ecosystems
- Vol. 7 (1) , 1-18
Innovative deep learning artificial intelligence applications for predicting relationships between individual tree height and diameter at breast height
- 3 April 2020
- Published by Elsevier in Forest Ecosystems
- Vol. 7 (1) , 1-18
Open AccessInnovative deep learning artificial intelligence applications for predicting relationships between individual tree height and diameter at breast height
- Open Access
- 1 April 2020
- Published by Sistema de Bibliotecas UACH in Bosque (Valdivia)
- Vol. 41 (1) , 25-34
Artificial neural network models for predicting relationships between diameter at breast height and stump diameter: Crimean pine stands at ÇAKÜ Forest
This study introduces the artificial neural networks (ANN) function to model relationship between diameter at breast height (dbh) and stump diameter and investigates the potential of ANN model to obtain the prediction of dbh. In total, 583 diameters at breast height-stump diameter pairs were measured in 61 plots sampled from Crimean pine [Pinus nigra subsp. pallasiana] stands in ÇAKÜ Research Forest, Çankırı, Turkey. The network models, including the activation functions of function between input layer and hidden layer and pure-lin function between hidden layer and output layer (A6 alternative) with 12 # neurons, were found to the better predictive with lower error values including SSE (2585.3869), AIC (821.5731), BIC (825.7817), RMSE (2.2831), MSE (5.2125) and higher fitting value, such as R2 adj (0.9372), than those of other prediction methods. The best predictive ANN model resulted in the reductions of SSE, AIC, BIC, RMSE and MSE by 9.8486 %, 5.9018 %, 5.8735 %, 5.0519 % and 9.8486 %, and R2 adj in the increase of 0.7377 % as compared to those by the regression model. This present study has underlined the capability of the ANN model for predicting the relationship between dbh and stump diameter. This novel artificial intelligence technique provides a modeling alternative for forest managers to predict dbh required information for the management of forests.
- 1 April 2020
- Published by Sistema de Bibliotecas UACH in Bosque (Valdivia)
- Vol. 41 (1) , 25-34
Open AccessArtificial neural network models for predicting relationships between diameter at breast height and stump diameter: Crimean pine stands at ÇAKÜ Forest
This study introduces the artificial neural networks (ANN) function to model relationship between diameter at breast height (dbh) and stump diameter and investigates the potential of ANN model to obtain the prediction of dbh. In total, 583 diameters at breast height-stump diameter pairs were measured in 61 plots sampled from Crimean pine [Pinus nigra subsp. pallasiana] stands in ÇAKÜ Research Forest, Çankırı, Turkey. The network models, including the activation functions of function between input layer and hidden layer and pure-lin function between hidden layer and output layer (A6 alternative) with 12 # neurons, were found to the better predictive with lower error values including SSE (2585.3869), AIC (821.5731), BIC (825.7817), RMSE (2.2831), MSE (5.2125) and higher fitting value, such as R2 adj (0.9372), than those of other prediction methods. The best predictive ANN model resulted in the reductions of SSE, AIC, BIC, RMSE and MSE by 9.8486 %, 5.9018 %, 5.8735 %, 5.0519 % and 9.8486 %, and R2 adj in the increase of 0.7377 % as compared to those by the regression model. This present study has underlined the capability of the ANN model for predicting the relationship between dbh and stump diameter. This novel artificial intelligence technique provides a modeling alternative for forest managers to predict dbh required information for the management of forests.