{"id":"W3184069529","doi":"10.1016/j.jhazmat.2021.126746","title":"Towards a better understanding of deep convolutional neural network processes for recognizing organic chemicals of environmental concern","year":2021,"lang":"en","type":"article","venue":"Journal of Hazardous Materials","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"Environment and Climate Change Canada; Concordia University","funders":"","keywords":"Convolutional neural network; Categorization; Artificial intelligence; Computer science; Pattern recognition (psychology); Redundancy (engineering); Principal component analysis; Process (computing); Machine learning","routes":{"ca_aff":true,"ca_fund":false,"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":[0.0003500197,0.0005897923,0.0003979488,0.0003550822,0.0002333211,0.00115969,0.0008784981,0.0009393583,0.002002923],"category_scores_gemma":[0.001054578,0.0004057917,0.0005447224,0.0002719163,0.0004070501,0.00198697,0.0005891987,0.001893155,0.0005126748],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008268034,"about_ca_system_score_gemma":0.001118214,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009741576,"about_ca_topic_score_gemma":0.01137439,"domain_scores_codex":[0.9999249,0.00001271722,0.000004622388,0.00002110446,0.0000209544,0.00001560374],"domain_scores_gemma":[0.9997312,0.0001096179,0.00003181673,0.00004447255,0.00006620446,0.0000165704],"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":[0.00006910803,0.0001457584,0.00259831,0.000175466,0.00008789424,0.00009198711,0.00007644633,0.7900746,0.03659619,0.07649158,0.00221214,0.09138048],"study_design_scores_gemma":[0.000001328917,0.000006243778,0.0001849013,0.000004364646,0.000003789835,0.000007000793,0.00000446645,0.9877191,0.001625437,0.009931974,0.0005089852,0.000002489714],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08454113,0.001494945,0.9050397,0.001531633,0.00009407575,0.00003526418,0.0005177584,0.0006895386,0.006055999],"genre_scores_gemma":[0.8343266,0.002526848,0.1527,0.0004223605,0.00008073299,0.00006067718,0.0007900505,0.0001018692,0.008990858],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009741576,"threshold_uncertainty_score":0.01936972,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04995128654637926,"score_gpt":0.287174677872877,"score_spread":0.2372233913264978,"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."}}