{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001376901,0.0001184636,0.0005722232,0.0001122054,0.0003514942,0.00002597187,0.00003699248,0.00003148877,0.000006963191],"category_scores_gemma":[0.000274734,0.00009858506,0.0001200892,0.0002183184,0.0002668779,0.0001656963,0.0000225362,0.0004986892,3.155769e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005663582,"about_ca_system_score_gemma":0.00005236237,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009129274,"about_ca_topic_score_gemma":0.000004099235,"domain_scores_codex":[0.9981431,0.0004374794,0.0007389906,0.0002103094,0.0003406696,0.0001294919],"domain_scores_gemma":[0.9981987,0.0006747472,0.0007201154,0.00004596106,0.000236083,0.0001244068],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.05978145,0.01076851,0.07718378,0.002853753,0.001099876,0.0000365606,0.001637097,0.0002217851,0.002415485,0.002172773,0.001122266,0.8407066],"study_design_scores_gemma":[0.09674679,0.127685,0.5559799,0.001773388,0.002203431,0.005499429,0.07891047,0.08979762,0.0005756128,0.01194807,0.02770239,0.001177947],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9485872,0.008855524,0.03725526,0.004165065,0.0001433155,0.0007006649,0.0002221704,0.00003220096,0.00003854752],"genre_scores_gemma":[0.9469432,0.0005447649,0.05122507,0.0002013611,0.0005635799,0.00008990033,0.0004114088,0.00001489327,0.000005778073],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8395287,"threshold_uncertainty_score":0.4020182,"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."}}