Incorporating phylogenetic conservatism and trait collinearity into machine learning frameworks can better predict macroinvertebrate traits DOI Creative Commons
Shuyin Li, Qingyi Luo,

Ruiwen Li

et al.

Frontiers in Ecology and Evolution, Journal Year: 2023, Volume and Issue: 11

Published: Aug. 29, 2023

In the face of rapid environmental changes, understanding and monitoring biological traits functional diversity are crucial for effective biomonitoring. However, when it comes to freshwater macroinvertebrates, a significant dearth trait data poses major challenge. this opinion article, we put forward machine-learning framework that incorporates phylogenetic conservatism collinearity, aiming provide better vision predicting macroinvertebrate in ecosystems. By adopting proposed framework, can advance biomonitoring efforts Accurate predictions enable us assess diversity, identify stressors, monitor ecosystem health more effectively. This information is vital making informed decisions regarding conservation management strategies, especially context rapidly changing environments.

Language: Английский

Comparación de métodos de aprendizaje automático para predicción de valores de cría genómicos en características de crecimiento en bovinos Suizo Europeo DOI Creative Commons

José Luis Vélez Labrada,

Paulino Pérez‐Rodríguez, Mohammad Ali Nilforooshan

et al.

Revista Mexicana de Ciencias Pecuarias, Journal Year: 2025, Volume and Issue: 16(1), P. 179 - 193

Published: Feb. 19, 2025

Los algoritmos de Aprendizaje Automático (AA) han demostrado ventaja al abordar desafíos asociados con la cantidad y complejidad información, permiten descubrir patrones, realizar análisis eficientes servir como herramienta para toma decisiones. Este estudio, tuvo objetivo comparar cuatro métodos AA: redes neuronales artificiales (RN), árboles regresión (AR), bosques aleatorios (BA) máquina soporte vectorial (SVM) predecir el valor genómico en bovinos Suizo Europeo utilizando registros fenotípicos pesos nacimiento (PN), destete (PD) año (PA), así información genómica. resultados indican que capacidad predictiva los modelos varía según característica disponible. En general, RN, BA SVM mostraron un desempeño similar, mientras AR inferior. La metodología destacó mayor potencial, obteniendo valores más altos correlación Pearson entre fenotipos corregidos genéticos predichos PD. A pesar costo computacional, RN razonable, especialmente PN PA. selección del modelo final depende las necesidades particulares aplicación, factores prácticos disponibilidad datos, recursos computacionales interpretabilidad; pero surgieron opciones sólidas varias categorías.

Citations

0
Comparative analysis of genomic prediction models based on body weight trait in large yellow croaker (Larimichthys crocea) DOI

Jialu Fang,

Qinglei Xu, Limin Feng

et al.

Aquaculture, Journal Year: 2025, Volume and Issue: 599, P. 742125 - 742125

Published: Jan. 7, 2025

Language: Английский

Citations

0
Enhancing fish farmers’ welfare through digital agricultural innovation platforms: Evidence from Nigeria DOI Creative Commons
Abraham Falola, Ridwan Mukaila, A.S. Olanrewaju

et al.

Ege Üniversitesi Ziraat Fakültesi Dergisi, Journal Year: 2024, Volume and Issue: 61(3), P. 273 - 283

Published: June 1, 2024

Objective: Despite the potential of digital innovation platforms to improve farmers’ welfare its adoption is low due scanty empirical evidence on subject matter. Therefore, this study examined contribution fish farm output and income in Nigeria. Material Methods: Data gathered from 187 catfish farmers were analysed using descriptive statistics t-tests. Results: The results revealed that platform provides credit facilities boost their production activities. Before adopted platform, they stocked an average 733.12 fingerlings, which increased 952.83 fingerlings after adopting innovation. Also, significantly 742.28 kg 1,057.81 kg. Fish revenue farming consequently N540,905.11 Nigerian Naira(USD 1,307.01) N780,444.98 1,885.82) by 42.51% 44.29%, respectively. Conclusion: Digital improved welfare. Based this, advocates should be encouraged adopt creating awareness providing more funds through platforms.

Language: Английский

Citations

1
Breeding evaluations in aquaculture using neural networks DOI Creative Commons
Christos Palaiokostas

Aquaculture Reports, Journal Year: 2024, Volume and Issue: 39, P. 102468 - 102468

Published: Nov. 15, 2024

Language: Английский

Citations

1
Incorporating phylogenetic conservatism and trait collinearity into machine learning frameworks can better predict macroinvertebrate traits DOI Creative Commons
Shuyin Li, Qingyi Luo,

Ruiwen Li

et al.

Frontiers in Ecology and Evolution, Journal Year: 2023, Volume and Issue: 11

Published: Aug. 29, 2023

In the face of rapid environmental changes, understanding and monitoring biological traits functional diversity are crucial for effective biomonitoring. However, when it comes to freshwater macroinvertebrates, a significant dearth trait data poses major challenge. this opinion article, we put forward machine-learning framework that incorporates phylogenetic conservatism collinearity, aiming provide better vision predicting macroinvertebrate in ecosystems. By adopting proposed framework, can advance biomonitoring efforts Accurate predictions enable us assess diversity, identify stressors, monitor ecosystem health more effectively. This information is vital making informed decisions regarding conservation management strategies, especially context rapidly changing environments.

Language: Английский

Citations

0