{"id":"W6940664403","doi":"10.1021/jacs.4c06595.s001","title":"Machine\\nLearning in Complex Organic Mixtures: Applying\\nDomain Knowledge Allows for Meaningful Performance with Small Data\\nSets","year":2024,"lang":"en","type":"article","venue":"Figshare","topic":"Mycorrhizal Fungi and Plant Interactions","field":"Agricultural and Biological Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Leverage (statistics); Domain (mathematical analysis); Domain knowledge; Knowledge-based systems; Complex system","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002770894,0.001663326,0.001018341,0.0009397773,0.0007168886,0.001679783,0.001345039,0.001396095,0.002627414],"category_scores_gemma":[0.005828596,0.000507594,0.001204542,0.0009382595,0.0008550093,0.002000559,0.001504783,0.002615834,0.001404011],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007172747,"about_ca_system_score_gemma":0.001259428,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008637466,"about_ca_topic_score_gemma":0.01296848,"domain_scores_codex":[0.9992552,0.0002916804,0.00005177634,0.0002407586,0.00008935815,0.00007119471],"domain_scores_gemma":[0.9972248,0.002177811,0.0001000093,0.0002662067,0.0001714143,0.00005981393],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004779208,0.000252376,0.005285557,0.0003008382,0.0003000912,0.0001270699,0.0001454154,0.7398164,0.01182636,0.003115459,0.002696637,0.2356559],"study_design_scores_gemma":[0.0000138784,0.00004882273,0.0007715012,0.00001796998,0.00001591698,0.0000176823,0.00003877711,0.9893439,0.004434859,0.003905876,0.001375751,0.00001499772],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1564229,0.002849534,0.8267752,0.001325761,0.0002081011,0.0002146992,0.001133618,0.006177752,0.004892574],"genre_scores_gemma":[0.6856446,0.0009521586,0.304833,0.0005810394,0.0001316007,0.0002417458,0.003892066,0.0003327434,0.003391143],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008637466,"threshold_uncertainty_score":0.01717436,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08637972606285851,"score_gpt":0.25704587871023,"score_spread":0.1706661526473715,"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."}}