Workflow of the proposed enhanced Random Forest (eRF) algorithm for volcanic lithology classification.

<p>Data preprocessing includes quality control, Z-score standardization, and targeted imbalance correction with Borderline-SMOTE. The eRF is trained under nested GroupKFold cross-validation (inner loop: hyperparameter tuning by macro-F1; outer loop: model assessment). Trees adopt C4.5 gain-rat...

Full description

Saved in:

Bibliographic Details
Main Author:	Xiu Jin (9452603) (author)
Other Authors:	Taiji Yu (22574368) (author), Pujun Wang (8278896) (author), Minglong Nie (22574371) (author), Shichang Lu (20290999) (author)
Published:	2025
Subjects:	Cell Biology Biotechnology Cancer Biological Sciences not elsewhere classified Information Systems not elsewhere classified wangfu fault depression rla5 )&# 8212 remains challenging due propagation neural network prioritizing robust predictors optimized ensemble provides kendall &# 8217 5 decision trees resolution lithology logs level feature selection informed feature optimization correlated continuous logs limited core control enhanced random forest complex volcanic sequences erf synergistically integrates erf attains 96 bpnn ), k geologically significant facies encompassing 18 lithologies 18 lithologies stabilize feature conventional logs volcanic reservoirs enhanced machine core innovation svm ), knn ), ac ), vector machine uncored intervals transferable route trachytic tuff together accounting standard rf spot prediction specifically engineered songliao basin ratio splitting precision classification overall accuracy optimize node nearest neighbors model decisions learning framework huoshiling formation geologically attuned gamma ray extreme heterogeneity centric challenges
Tags:	Add Tag No Tags, Be the first to tag this record!

Workflow of the proposed enhanced Random Forest (eRF) algorithm for volcanic lithology classification.

Similar Items