{"id":"W7106002583","doi":"10.7910/dvn/0c1i9w","title":"Supplement to the Oil and Gas Extraction Data in OGIM","year":2025,"lang":"","type":"dataset","venue":"Harvard Dataverse","topic":"","field":"","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Fossil fuel; Extraction (chemistry); Data extraction; Petroleum; Space (punctuation)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science","research_integrity","insufficient_payload"],"consensus_categories":["metaepi_narrow","open_science","insufficient_payload"],"category_scores_codex":[0.006586041,0.001626121,0.001372471,0.001599304,0.0007501717,0.001337199,0.008541289,0.0006713379,0.06696431],"category_scores_gemma":[0.003945152,0.001476669,0.0001267264,0.002196066,0.0004182549,0.00281701,0.02248385,0.002470928,0.3317681],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001161853,"about_ca_system_score_gemma":0.001325099,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.01072842,"about_ca_topic_score_gemma":0.06322248,"domain_scores_codex":[0.9882439,0.001515744,0.002111965,0.004329953,0.00195487,0.001843599],"domain_scores_gemma":[0.9760452,0.001032505,0.000878491,0.02116725,0.0002033814,0.0006731607],"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":[0.001137951,0.0007865023,0.00005316909,0.000731879,0.0003643882,0.0004992169,0.0001889496,0.00006582254,0.0001154451,0.00006693315,0.9816209,0.01436881],"study_design_scores_gemma":[0.002698923,0.0001436041,0.0001995938,0.001382982,0.001291954,0.0001050293,0.001502064,0.001007258,0.00002504929,0.00001118999,0.9903156,0.001316816],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00006755426,0.00001819096,0.00006635865,0.0007543762,0.003058777,0.001316669,0.9943818,0.00006670799,0.0002695962],"genre_scores_gemma":[0.00003080301,0.01330302,0.001249685,0.001320555,0.0007326491,0.0003041265,0.9811542,0.0001183572,0.001786596],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.2648038,"threshold_uncertainty_score":0.9998304,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03268663765964462,"score_gpt":0.312006691020394,"score_spread":0.2793200533607493,"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."}}