{"id":"W4312161061","doi":"10.1016/j.ajcnut.2022.11.022","title":"Natural language processing and machine learning approaches for food categorization and nutrition quality prediction compared with traditional methods","year":2022,"lang":"en","type":"article","venue":"American Journal of Clinical Nutrition","topic":"Nutritional Studies and Diet","field":"Medicine","cited_by":58,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"Canadian Institutes of Health Research","keywords":"Categorization; Computer science; Quality (philosophy); Artificial intelligence; Machine learning; Natural language processing","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":[0.002693421,0.0005494475,0.0005657087,0.001714226,0.0004962242,0.001743539,0.0007994507,0.000673436,0.003715465],"category_scores_gemma":[0.005734524,0.0001475348,0.0008338626,0.001434655,0.000383948,0.002404602,0.0008092024,0.001266353,0.001094802],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007330609,"about_ca_system_score_gemma":0.001060954,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00678245,"about_ca_topic_score_gemma":0.006573737,"domain_scores_codex":[0.9985508,0.0006407429,0.0001060348,0.0003394288,0.0002995096,0.00006337219],"domain_scores_gemma":[0.9939011,0.004854449,0.0002190112,0.0003043585,0.0006317363,0.00008936365],"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":[0.001565555,0.001047739,0.0139398,0.0003535692,0.0003028518,0.0001525362,0.0002734267,0.02780008,0.01312887,0.005264361,0.003745676,0.9324255],"study_design_scores_gemma":[0.0001177634,0.0004233701,0.01579018,0.00005043557,0.0002360787,0.0001857575,0.000424192,0.9515801,0.007387125,0.01959758,0.004159645,0.00004784347],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.316148,0.004427475,0.6626119,0.001968422,0.0005243383,0.0004828639,0.001670565,0.002255487,0.009910947],"genre_scores_gemma":[0.6846685,0.001418178,0.3053261,0.0005401085,0.0003872263,0.0002804932,0.001794572,0.0001158821,0.005469024],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00678245,"threshold_uncertainty_score":0.01424432,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1599924280321632,"score_gpt":0.4153712265547376,"score_spread":0.2553787985225744,"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."}}