A Comparative Study Of 11 Non-linear Regression Models Highlighting Autoencoder, Dbn, And Svr, Enhanced By Shap Importance Analysis In Soybean Branching Prediction

This article discusses the use of artificial intelligence in predicting soybean branching through a comparison of 11 non-linear regression models. The study found that Support Vector Regression, Polynomial Regression, DBN, and Autoencoder were the most accurate models for phenotype prediction. Additionally, the article highlights the use of feature importance and gene ontology enrichment in the assessment of deep learning approaches.