Application of Artificial Neural Networks to predict the final fruit weight and random forest to select important variables in native population of melon (Cucumis melo L.)

Rad, Mohammad Reza Naroui; Koohkan, Shirali; Fanaei, Hamid Reza; Rad, Mohammad Reza Pahlavan

doi:10.1016/j.scienta.2014.10.025

Cited by 59 publications

(22 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Knowledge on genetic variability can help breeders to select parental genotypes for hybridization leading to new cultivars. Studies have been conducted in order to characterize gene bank collections of melon germplasm (Naroui Rad et al 2015;Naroui Rad et al 2014). Using indirect selection criteria to select high yielding genotypes includes the knowledge of component traits (Aparicio et al 2000).…”

Section: Introductionmentioning

confidence: 99%

Evaluation of Melon (Cucumis Melo. L) Genotypes Aiming Effective Selection of Parents for Breeding Directed at High Yield under Drought Stress Condition

Rad

Ghasemi

Koohpayegani

2017

Journal of Horticultural Research

Self Cite

View full text Add to dashboard Cite

This study examined 36 accessions of melon grown originally in dry and warm zones managed in the gene bank in the Institute of Seed and Plant Improvement, Karaj. To characterize and compare results of the field experiment statistical procedures, hierarchical cluster analysis, principal component analysis (PCA), and multiple linear regressions were applied. The results indicated a high degree of variability for single plant yield, fruit weight, fruit length, and cavity diameter. In addition, the variables of fruit size displayed a high differentiating power. Positive and negative correlation coefficients were identified among the studied traits as well. The highest significant positive correlation was found between the plant yield and fruit weight (0.88). Also, the highest significant negative correlation was observed between relative water content and canopy temperature (−0.58). Melon accessions were classified into three main clusters, indicating that the characterized melon collection has a high potential for particular breeding goals. The usefulness of the identified correlations among traits is valuable for the potential use in breeding projects directed for fruit size and yielding in dry condition. Multiple linear regressions were done and R 2 (Coefficient of determination) and RMSE (Root Mean Squared Error) were 0.97 and 229, respectively.

show abstract

Section: Introductionmentioning

confidence: 99%

Evaluation of Melon (Cucumis Melo. L) Genotypes Aiming Effective Selection of Parents for Breeding Directed at High Yield under Drought Stress Condition

Rad

Ghasemi

Koohpayegani

2017

Journal of Horticultural Research

Self Cite

View full text Add to dashboard Cite

show abstract

“…In most applications, the neuron number is determined by trial and error [84]. The different topology arrays resulting from these combinations were evaluated in the MATLAB neural network toolbox [85], and the following transfer functions were used: tangent sigmoidal hyperbolic (tansig), logarithmic sigmoidal hyperbolic (logsig), and pure lineal (purelin) [85,86], with a learning rate of 0.5 [63], 1000 epochs [64], minimum performance gradient of 1e −07 and adaptation value of 0.001. The tangent sigmoidal hyperbolic (tansig) transfer function presented the best performance in the hidden layers, and the pure lineal (purelin) transfer function presented the best performance in the output layers, defined by its lower mean square error (MSE) values for the substrate and soil (Table 1).…”

Section: Neuron Topologies In the Hidden Layersmentioning

confidence: 99%

“…The learning rate of 0.5 used in this research for the substrate and the soil culture systems is similar to the learning rate (0.6) used in [63] and is in the range of the recommended values (0.05 to 0.5 [96], 0.1 to 0.7 [97], and 0.05 to 0.75 [98]), where the learning rate value has no influence on the ANN error [97].…”

Section: Training Validation and Test Processes Of The Annsmentioning

confidence: 99%

“…It was also observed that increasing the layers did not decrease the MSE, which began to remain constant at 10, 7, and 5 neurons in the hidden layers, with eight epochs for the substrate and 10, 8, and 5 neurons in the hidden layers, with six epochs for the soil. The study in [63] evaluated a three-layer ANN with different neurons in the hidden layer and determined different changes in the mean prediction error as the topology increased-that is, a reduction from 3.34% to 2.21% (9-1-1 to 9-5-1), and an increment from 2.21% to 2.56% (9-5-1 to 9-9-1).…”

Section: Training Validation and Test Processes Of The Annsmentioning

confidence: 99%

See 1 more Smart Citation

Artificial Neural Network Modeling of Greenhouse Tomato Yield and Aerial Dry Matter

et al. 2020

View full text Add to dashboard Cite

Non-linear systems, such as biological systems, can be simulated by artificial neural network (ANN) techniques. This research aims to use ANN to simulate the accumulated aerial dry matter (leaf, stem, and fruit) and fresh fruit yield of a tomato crop. Two feed-forward backpropagation ANNs, with three hidden layers, were trained and validated by the Levenberg–Marquardt algorithm for weights and bias adjusted. The input layer consisted of the leaf area, plant height, fruit number, dry matter of leaves, stems and fruits, and the growth degree-days at 136 days after transplanting (DAT); these were obtained from a tomato crop, a hybrid, EL CID F1, with indeterminate growth habits, grown with a mixture of peat moss and perlite 1:1 (v/v) (substrate) and calcareous soil (soil). Based on the experimentation of the ANNs with one, two and three hidden layers, with MSE values less than 1.55, 0.94 and 0.49, respectively, the ANN with three hidden layers was chosen. The 7-10-7-5-2 and 7-10-8-5-2 topologies showed the best performance for the substrate (R = 0.97, MSE = 0.107, error = 12.06%) and soil (R = 0.94, MSE = 0.049, error = 13.65%), respectively. These topologies correctly simulated the aerial dry matter and the fresh fruit yield of the studied tomato crop.

show abstract

“…For example, Hughes et al (1996);Castro-Tanzi et al (2014); Di Paola et al (2016) and Rad et al (2015) applied predictive models for crop yield or crop production, which can be considered as an output in the agrarian sector. Thacher et al (1996); Davis and Lopez-Carr (2014) and van der Sluis et al (2016) dealt with the prediction of economic characteristics, such as the use of soil.…”

Section: Predictive Modelmentioning

confidence: 99%

The impact of investment support from the Rural Development Programme of the Czech Republic for 2007-2013 on the economic efficiency of farms

Hlavsa¹,

Hruška²,

Туркова³

2017

Stud. Agr. Econ.

View full text Add to dashboard Cite

Application of Artificial Neural Networks to predict the final fruit weight and random forest to select important variables in native population of melon (Cucumis melo L.)

Cited by 59 publications

References 14 publications

Evaluation of Melon (Cucumis Melo. L) Genotypes Aiming Effective Selection of Parents for Breeding Directed at High Yield under Drought Stress Condition

Evaluation of Melon (Cucumis Melo. L) Genotypes Aiming Effective Selection of Parents for Breeding Directed at High Yield under Drought Stress Condition

Artificial Neural Network Modeling of Greenhouse Tomato Yield and Aerial Dry Matter

The impact of investment support from the Rural Development Programme of the Czech Republic for 2007-2013 on the economic efficiency of farms

Contact Info

Product

Resources

About