{"id":"W4398805385","doi":"10.7910/dvn/sovpa4","title":"Replication Data for: Multi-label Prediction for Political Text-as-Data","year":2021,"lang":"en","type":"dataset","venue":"London School of Economics and Political Science Theses Online (London School of Economics and Political Science)","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Replication (statistics); Computer science; Computational biology; Artificial intelligence; Biology; Virology","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":[0.004384549,0.002183442,0.0009915036,0.003078575,0.001599953,0.002359601,0.003946123,0.002452153,0.0363644],"category_scores_gemma":[0.01816616,0.0007272076,0.001382148,0.004642958,0.0009823106,0.002709158,0.002999836,0.00331004,0.06578612],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00190676,"about_ca_system_score_gemma":0.002478027,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02297685,"about_ca_topic_score_gemma":0.04560216,"domain_scores_codex":[0.9964638,0.001095587,0.0004069188,0.0008110171,0.0009324615,0.0002901923],"domain_scores_gemma":[0.9890676,0.00241809,0.0007071651,0.004800754,0.002350483,0.0006558603],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001409994,0.0001211339,0.001769016,0.0003427919,0.00003834475,0.00006287923,0.00007117809,0.0007198955,0.0003260099,0.001057129,0.9881535,0.00719717],"study_design_scores_gemma":[0.0006420622,0.00009601517,0.008743748,0.000220715,0.00003694397,0.0002308014,0.0003486948,0.007912727,0.00263465,0.00462595,0.974414,0.00009363735],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00193439,0.0001430164,0.001259038,0.0005153828,0.0002336735,0.0001271583,0.9906965,0.003048049,0.002042683],"genre_scores_gemma":[0.002293192,0.00004264454,0.002776117,0.00007991591,0.00002398081,0.0002501006,0.9930883,0.0001909618,0.001254719],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.0363644,"threshold_uncertainty_score":0.1216511,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0807064995258494,"score_gpt":0.3732060154628786,"score_spread":0.2924995159370292,"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."}}