{"id":"W1971156133","doi":"10.1371/journal.pone.0006777","title":"Parsing Social Network Survey Data from Hidden Populations Using Stochastic Context-Free Grammars","year":2009,"lang":"en","type":"article","venue":"PLoS ONE","topic":"HIV, Drug Use, Sexual Risk","field":"Medicine","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of Allergy and Infectious Diseases; National Institute of Nursing Research; Canadian Institutes of Health Research; University of California, San Diego; Center for AIDS Research, University of Washington; National Institute on Drug Abuse; U.S. Public Health Service; Centre de Coopération Internationale en Recherche Agronomique pour le Développement","keywords":"Parsing; Computer science; Context-free grammar; Rule-based machine translation; Context (archaeology); Natural language processing; Biology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004714562,0.0002142631,0.0005740665,0.00007469553,0.0003153455,0.0000558537,0.000475729,0.0001507593,0.00009889097],"category_scores_gemma":[0.001582979,0.0002293086,0.00004253762,0.0003650168,0.0000752859,0.0002094111,0.0002106989,0.0003619039,0.00007551516],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000135903,"about_ca_system_score_gemma":0.000131871,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002305124,"about_ca_topic_score_gemma":0.002957758,"domain_scores_codex":[0.9977935,0.0002017097,0.0004326797,0.0004775919,0.0006104329,0.000484082],"domain_scores_gemma":[0.9979833,0.0003044138,0.0001629817,0.001202714,0.0001493092,0.0001972387],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.004665101,0.02038167,0.6318196,0.0003199542,0.005810078,0.0004499015,0.02678178,0.001651773,0.04137579,0.003301855,0.1201488,0.1432937],"study_design_scores_gemma":[0.005940376,0.000424085,0.8514808,0.001407381,0.003391565,0.00001155513,0.002265922,0.1092995,0.0001184752,0.02457574,0.00005947176,0.001025044],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9910005,0.0004326765,0.005300011,0.001301879,0.0001478672,0.0005628394,0.0008242131,0.0001777252,0.0002523384],"genre_scores_gemma":[0.9743757,0.000005372771,0.01996849,0.0005570879,0.001667337,0.000002865806,0.003240066,0.00004866572,0.0001344652],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2196612,"threshold_uncertainty_score":0.9350935,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4193035123806864,"score_gpt":0.3882867108926178,"score_spread":0.03101680148806862,"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."}}