{"id":"W2540949022","doi":"10.2139/ssrn.2855977","title":"Using Machine Learning to Predict Outcomes in Tax Law","year":2016,"lang":"en","type":"article","venue":"SSRN Electronic Journal","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":45,"is_retracted":false,"has_abstract":false,"ca_institutions":"Queen's University; Vector Institute; Royal Ontario Museum; University of Toronto","funders":"","keywords":"Artificial intelligence; Machine learning; Field (mathematics); Computer science; Government (linguistics); Analytics; Set (abstract data type); Data science","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00342943,0.0001159534,0.0001812635,0.0001273759,0.0006367716,0.00007377104,0.0003743168,0.00008406153,0.0001880539],"category_scores_gemma":[0.0006850773,0.00008274374,0.00008156288,0.0002741431,0.0001848965,0.0003833703,0.00004825381,0.0009335387,0.0001178013],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001910594,"about_ca_system_score_gemma":0.001153577,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.01224378,"about_ca_topic_score_gemma":0.2495207,"domain_scores_codex":[0.9965575,0.0003257679,0.0003202508,0.0001834121,0.0004215934,0.002191442],"domain_scores_gemma":[0.9993836,0.0001836853,0.000111347,0.00009725983,0.00007314129,0.0001509747],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00003026608,0.00004442106,0.1372015,7.514304e-7,0.00003088654,0.000007536269,0.003214853,0.0002675283,0.002196041,0.8357554,0.000008890927,0.02124197],"study_design_scores_gemma":[0.0006986794,0.0007081643,0.001068409,0.0002067835,0.00004366247,0.0000921849,0.02437995,0.0005256542,0.002090288,0.8350222,0.1343129,0.0008511867],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9633835,0.0003696052,0.01818983,0.009298745,0.0005399351,0.0002528124,0.00000309183,0.00007703661,0.007885439],"genre_scores_gemma":[0.9944862,0.0004165894,0.0001914072,0.0001755392,0.0003156606,0.00000290054,1.422372e-7,0.00001803537,0.004393593],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2372769,"threshold_uncertainty_score":0.9943338,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04042502317634256,"score_gpt":0.356381636546554,"score_spread":0.3159566133702114,"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."}}