{"id":"W4390948539","doi":"10.1039/d3dd00199g","title":"Reproducibility in materials informatics: lessons from ‘A general-purpose machine learning framework for predicting properties of inorganic materials’","year":2024,"lang":"en","type":"article","venue":"Digital Discovery","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"Structural Genomics Consortium; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Advanced Scientific Computing Research; U.S. Department of Energy","keywords":"Informatics; Materials informatics; Computer science; Health informatics; Data science; Engineering; Engineering informatics; Medicine","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.04347269,0.0009450295,0.001557845,0.00199659,0.001596418,0.006841532,0.006683832,0.003884231,0.002511356],"category_scores_gemma":[0.07035348,0.0008490389,0.001846913,0.001627049,0.01987636,0.02106883,0.005158289,0.00934589,0.001374717],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001860926,"about_ca_system_score_gemma":0.003500646,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001736744,"about_ca_topic_score_gemma":0.001060311,"domain_scores_codex":[0.9914977,0.004076302,0.0005635545,0.001330978,0.00227161,0.0002598532],"domain_scores_gemma":[0.9317185,0.0427929,0.00119333,0.02027055,0.003066822,0.0009578561],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"qualitative","study_design_scores_codex":[0.00004291855,0.00009757857,0.001186633,0.0002794744,0.00005922042,0.00008744785,0.000677068,0.01373684,0.0007815389,0.8997647,0.006984416,0.07630214],"study_design_scores_gemma":[0.00001577515,0.00002527664,0.0002187877,0.00005847354,0.000007243018,0.00006000108,0.00005531413,0.02791113,0.001576678,0.9575893,0.01245346,0.00002851597],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01380945,0.004037046,0.9193363,0.04490899,0.0006459192,0.00007058171,0.0001638866,0.00199172,0.01503621],"genre_scores_gemma":[0.2867116,0.004220287,0.6996712,0.00316551,0.00108289,0.0002224963,0.0002838643,0.001086633,0.00355547],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9565273,"threshold_uncertainty_score":0.2299083,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02382068067150253,"score_gpt":0.2753458697987379,"score_spread":0.2515251891272353,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}