{"id":"W4387099797","doi":"10.54254/2755-2721/6/20230334","title":"Feature selection in text classification: Identifying spurious words with causal inference methods","year":2023,"lang":"en","type":"article","venue":"Applied and Computational Engineering","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Spurious relationship; Computer science; Inference; Causal inference; Artificial intelligence; Weighting; Feature selection; Feature (linguistics); Machine learning; Propensity score matching; Matching (statistics); Selection (genetic algorithm); Model selection; Selection bias; Pattern recognition (psychology); Data mining; Statistics; Mathematics","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":[0.0002769176,0.0001280277,0.000123292,0.0003207623,0.00007630647,0.0001549909,0.0002305647,0.00006154027,0.000001391889],"category_scores_gemma":[0.00002302788,0.0001283199,0.00001118578,0.001366623,0.00002010541,0.0002912786,0.00008847457,0.0001967555,0.00001078599],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000559779,"about_ca_system_score_gemma":0.00005040384,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003954846,"about_ca_topic_score_gemma":0.000002612994,"domain_scores_codex":[0.9991033,0.00001928366,0.0001458142,0.0003474449,0.0001961604,0.0001879792],"domain_scores_gemma":[0.9994978,0.0002188646,0.00005758259,0.000132226,0.00004657495,0.00004698578],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00000843852,0.00002307445,0.000846001,0.00006483166,0.00002056885,0.000006944329,0.0006369292,0.193229,0.01728277,0.6928881,0.0002842916,0.0947091],"study_design_scores_gemma":[0.0001921619,0.0000142997,0.1159462,0.00002997026,0.000002322669,0.00001688843,0.00003437409,0.8762512,0.0009103737,0.005542979,0.0008794189,0.0001798155],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006132029,0.00002384719,0.9922374,0.0005123335,0.000043712,0.0001493224,0.000001665195,0.0006774468,0.0002221771],"genre_scores_gemma":[0.6011531,0.00000970028,0.3986371,0.00003604238,0.00002136062,0.00007948544,0.00002583158,0.000008742597,0.00002868108],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6873451,"threshold_uncertainty_score":0.5232735,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02398821052233942,"score_gpt":0.3065342753834491,"score_spread":0.2825460648611097,"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."}}