Have a personal or library account? Click to login

Effect of vermicompost application on the development of plant properties and root architecture analysis with machine learning in Buxus herlandii

Open Access
|May 2025

Figures & Tables

Figure 1.

The MLP structure with seven inputs, five outputs, and five hidden neurons. MLP, multilayer perceptron.
The MLP structure with seven inputs, five outputs, and five hidden neurons. MLP, multilayer perceptron.

Figure 2.

Change rates of plant upper part characteristics compared to the control values of the analysis results after vermicompost applications.
Change rates of plant upper part characteristics compared to the control values of the analysis results after vermicompost applications.

Figure 3.

Effect of vermicompost applications on leaf chlorophyll content.
Effect of vermicompost applications on leaf chlorophyll content.

Figure 4.

The rate of change of root architectural features compared to the control values of the analysis results after vermicompost applications.
The rate of change of root architectural features compared to the control values of the analysis results after vermicompost applications.

Figure 5.

The rate of change of nutrient content in roots.
The rate of change of nutrient content in roots.

Figure 6.

DT obtained by J.48 method. DT, decision tree.
DT obtained by J.48 method. DT, decision tree.

Figure 7.

Comparison of performance levels of models created using PART, J48, Multilayer Perceptron, and Multi Clas Classifier algorithms.
Comparison of performance levels of models created using PART, J48, Multilayer Perceptron, and Multi Clas Classifier algorithms.

Figure 8.

The predictive power of ML models. ML, machine learning.
The predictive power of ML models. ML, machine learning.

Formulas used in evaluation_

FormulaDefinition
1  TRP=TPiTPi+FNi×100%{\rm{TRP}} = {{{\rm{TPi}}} \over {{\rm{TPi}} + {\rm{FNi}}}} \times 100\% The TPR is the proportion of positive instances that are correctly classified by the model. Where TP is the number of true positive instances, and FN is the number of false negative instances. The larger the value the better
2  FPR=FPiFPi+TNi×100%{\rm{FPR}} = {{{\rm{FPi}}} \over {{\rm{FPi}} + {\rm{TNi}}}} \times 100\% FPR indicates the probability that a positive decision is wrong. The smaller the value, the better the performance of the model
3   Precision =TPiTPi+FPi×100{\rm{ Precision }} = {{{\rm{TPi}}} \over {{\rm{TPi}} + {\rm{FPi}}}} \times 100Precision is the ratio of the samples correctly predicted by the model to all the samples positively predicted
4  F Measure =2× Precision × Recall  Precision + Recall {\rm{F}} - {\rm{ Measure }} = {{2 \times {\rm{ Precision }} \times {\rm{ Recall }}} \over {{\rm{ Precision }} + {\rm{ Recall }}}}The F-measure is defined as the weighted harmonic mean of precision and recall
5   Accuracy =TPi+TNiTPi+TNi+FPi+FNi×100{\rm{ Accuracy }} = {{{\rm{TPi}} + {\rm{TNi}}} \over {{\rm{TPi}} + {\rm{TNi}} + {\rm{FPi}} + {\rm{FNi}}}} \times 100The accuracy of correct classification ranges between 0.5 and 1, where higher values indicate a better classifier. Accuracy values between 0.7 and 1 are generally considered acceptable

Effect of vermicompost application on plant upper part characteristics_

ApplicationsPlant height (cm)Number of shootsShoot length (mm)Leaf width (cm)Leaf length (cm)
Control11.5 ab7.7 c5.6 c7.7 c23.4 c
10 mL10.9 b11.2 a6.0 ab8.4 b26.4 ab
20 mL10.2 c8.7 b5.6 c9.6 a25.8 b
40 mL11.4 ab8.5 b7.0 a9.0 ab27.1 a
80 mL11.8 a7.4 c5.8 bc8.6 b25.8 b

Chemical properties of perlite_

Ingredients%Ingredients%
SiO271.0–75.0Cr0.0–0.1
AlO312.5–18.0Ba0.0–0.05
Na2O32.9–4.0PbO0.0–0.03/0.3
K2O0.5–5.0NiOTrace amount
CaO0.5–0.2CuTrace amount
Fe2O30.1–1.5BTrace amount
MgO0.02–0.5BeTrace amount
TiO20.03–0.2free silica0.0–0.2
MnO20.0–0.1Total chloridesTrace amount – 0.2
SO30.0–0.2Total sulphatesNone
FeO0.0–0.1

Effect of vermicompost application on root architectural properties_

ApplicationsRoot length (cm)Root surface area (cm2)Root volume (cm3)Root diameter (mm)Number of tipsNumber of forksNumber of crossings
Control35.60 ab2.90 a22 bc3 a17.99 ab6.77 a590 a
10 mL37.34 a2.57 b15 c2 b18.84 a6.37 ab572 ab
20 mL29.28 c2.85 a26 b3 a14.38 b5.19 c440 b
40 mL31.20 bc2.65 b31 a3 a14.62 b5.09 c449 b
80 mL33.46 b2.75 ab23 bc3 a14.10 b5.76 b470 b

Effect of vermicompost applications on root nutrient content_

ApplicationNPKCaMgFeCuZnMn
%mg · kg−1
Control2.99 b2075 a18991 c6309 a2876 a181 ab10.4 b58.4 a146 b
10 mL3.93 a1605 b20914 b6052 b2369 b160 b7.7 c31.3 c229 a
20 mL3.02 b1594 b24264 ab6386 a2744 ab224 a18.4 a41.8 bc118 c
40 mL3.73 ab1473 bc29988 a6200 ab1919 bc140 c8.6 c44.1 b123 bc
80 mL2.87 c1230 c25720 ab5068 c1739 c143 c15.0 ab43.5 b75.0 d
DOI: https://doi.org/10.2478/fhort-2025-0005 | Journal eISSN: 2083-5965 | Journal ISSN: 0867-1761
Language: English
Page range: 49 - 64
Submitted on: Sep 11, 2024
Accepted on: Mar 28, 2025
Published on: May 6, 2025
Published by: Polish Society for Horticultural Sciences (PSHS)
In partnership with: Paradigm Publishing Services
Publication frequency: 2 issues per year

© 2025 Ömer Sari, Elif Enginsu, Fisun Gürsel Çelikel, published by Polish Society for Horticultural Sciences (PSHS)
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.