{"id":"W4403808131","doi":"10.1039/d4dd00252k","title":"Agent-based learning of materials datasets from the scientific literature","year":2024,"lang":"en","type":"article","venue":"Digital Discovery","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund; University of Toronto","keywords":"Data science; Computer science","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":[],"consensus_categories":[],"category_scores_codex":[0.001819783,0.0005928029,0.0005273208,0.002272704,0.0006807846,0.00117816,0.001483235,0.0009727413,0.003138299],"category_scores_gemma":[0.006769963,0.0003032014,0.0009104555,0.001411362,0.0004001479,0.001550355,0.001491029,0.00110428,0.0008841311],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008957009,"about_ca_system_score_gemma":0.002043229,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004562703,"about_ca_topic_score_gemma":0.01470842,"domain_scores_codex":[0.99935,0.0002337776,0.00005333455,0.000151063,0.0001756573,0.00003617717],"domain_scores_gemma":[0.9971426,0.001726744,0.0001969823,0.0003065907,0.0004407599,0.0001862772],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001278248,0.002451416,0.02000728,0.001737732,0.0006235489,0.001258344,0.0007119496,0.2520336,0.01441356,0.0350889,0.08139453,0.5890009],"study_design_scores_gemma":[0.00008785999,0.00009878344,0.001528573,0.00007217508,0.00007079948,0.00007000448,0.0001413629,0.9573838,0.0062538,0.01569845,0.01857271,0.00002178178],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1972108,0.003045209,0.7350907,0.007275055,0.0007753149,0.001716748,0.01732136,0.01337592,0.02418886],"genre_scores_gemma":[0.4011395,0.0006645342,0.5728898,0.0007038325,0.0001072134,0.0008586517,0.01602232,0.0001916706,0.007422407],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004562703,"threshold_uncertainty_score":0.01049864,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0100241379096093,"score_gpt":0.2521744097452122,"score_spread":0.2421502718356029,"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."}}