2026-10-10T17:50:11Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33811943 2026-09-15T17:41:16Z category_272 category_39 category_734 category_931 category_106 portal_5 item_type_3 month_year_09_2026

<p>Evaluation metrics. The table shows the average evaluation metrics for XGBoost and two single-rule-based classifiers across the outer folds of the nested cross-validation. Baseline 1: single-rule-based classifier using the maximum obstacle height attempted (including unsuccessful attempts). Baseline 2: single-rule-based classifier using the maximum obstacle height successfully cleared (clear rounds only). AP: average precision; AUC: area under the ROC curve; ACC: accuracy; SEN: sensitivity; SPEC: specificity.</p> Marco Zanchi (24948749) Clara Bordin (18358329) Michela Ablondi (7402970) Vittoria Asti (14490794) Andrea Summer (10446731) Emanuela Valle (5531261) Laura Ozella (606377) Environmental Sciences not elsewhere classified Ecology Biological Sciences not elsewhere classified Information Systems not elsewhere classified Science Policy shapley additive explanations probability trajectories showed machine learning approaches extreme gradient boosting based talent identification based classifiers based approximately seven years ration &# 201 horse &# 8217 competition history expands identifying horses likely f &# 233 early competition trajectories level performance develops reach elite performance &# 233 competition records horses identified early prediction performance development performance data performance consistency model performance two single support evidence sport potential show jumping roc curve related progression questre internationale permutation importance obstacle difficulty maintaining attention level records influential predictors increasing separation findings demonstrate evaluation metrics considerable time classifier trained average precision average area analyses revealed analysed using 160 cm 0 %. <p>Evaluation metrics. The table shows the average evaluation metrics for XGBoost and two single-rule-based classifiers across the outer folds of the nested cross-validation. Baseline 1: single-rule-based classifier using the maximum obstacle height attempted (including unsuccessful attempts). Baseline 2: single-rule-based classifier using the maximum obstacle height successfully cleared (clear rounds only). AP: average precision; AUC: area under the ROC curve; ACC: accuracy; SEN: sensitivity; SPEC: specificity.</p> 2026-09-15T17:41:04Z Dataset Dataset 10.1371/journal.pone.0357799.t001 https://figshare.com/articles/dataset/_p_Evaluation_metrics_The_table_shows_the_average_evaluation_metrics_for_XGBoost_and_two_single-rule-based_classifiers_across_the_outer_folds_of_the_nested_cross-validation_Baseline_1_single-rule-based_classifier_using_the_maximum_obstacle_/33811943 CC BY 4.0