{"id":"W6976652103","doi":"10.60692/zybpg-91c05","title":"Leveraging feature selection to detect potential tax fraudsters","year":2020,"lang":"en","type":"article","venue":"Greater South Information System","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Taxpayer; Treasury; Payment; Enforcement; Evasion (ethics); Agency (philosophy); Feature (linguistics); Work (physics)","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000200972,0.0001713765,0.0001682181,0.0002342358,0.0001694339,0.0005800427,0.0006667284,0.0001003873,0.000005460842],"category_scores_gemma":[0.00002930491,0.0001600817,0.00005659187,0.0007335678,0.000009133603,0.001918632,0.0001837808,0.0001503938,0.000939851],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001466721,"about_ca_system_score_gemma":0.00003967563,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003271342,"about_ca_topic_score_gemma":4.220612e-8,"domain_scores_codex":[0.9986595,0.00006253672,0.0003800323,0.0002699032,0.0003771353,0.0002508922],"domain_scores_gemma":[0.9989958,0.000003357959,0.0002320164,0.0004094041,0.0001846779,0.0001747009],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004284781,0.00000934901,0.03794928,0.002254822,0.0003063821,0.00004509664,0.6929105,0.01118967,0.005208284,0.007360601,0.1308937,0.1114439],"study_design_scores_gemma":[0.002492343,0.0005379306,0.07668764,0.0004021922,0.0000506076,0.0003573073,0.009520726,0.7240569,0.1402814,0.00002038944,0.04351514,0.002077453],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01181788,7.801711e-7,0.9835651,0.001905035,0.0002686754,0.0004110629,0.00003077088,0.001425543,0.000575147],"genre_scores_gemma":[0.9320899,4.06928e-8,0.06547038,0.002210937,0.00009499035,0.00006515678,0.00001273412,0.000007836068,0.00004800732],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9202721,"threshold_uncertainty_score":0.9998381,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02713987545648489,"score_gpt":0.2097561702501647,"score_spread":0.1826162947936798,"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."}}