{"id":"W4390781999","doi":"10.1002/asi.24865","title":"<scp>Phenomenon‐based</scp> classification: An Annual Review of Information Science and Technology (ARIST) paper","year":2024,"lang":"en","type":"article","venue":"Journal of the Association for Information Science and Technology","topic":"Advanced Text Analysis Techniques","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Phenomenon; Identification (biology); Computer science; Discipline; Epistemology; Information system; Data science; Sociology; Social science; Engineering; Philosophy","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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.006534587,0.0001163661,0.0002540663,0.002891089,0.0004544151,0.0003683585,0.001653602,0.0001331791,7.365423e-7],"category_scores_gemma":[0.007601774,0.00008392557,0.00004804536,0.009416401,0.0008416761,0.02098752,0.0003171712,0.0002795896,0.00000311808],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005246239,"about_ca_system_score_gemma":0.001270173,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001182838,"about_ca_topic_score_gemma":9.09989e-7,"domain_scores_codex":[0.997376,0.00002428883,0.0009023381,0.000161211,0.001272155,0.0002639873],"domain_scores_gemma":[0.9908994,0.0001673938,0.001594087,0.0004297193,0.006841881,0.00006756313],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000002092739,0.00002888995,0.001405105,0.0004766998,0.00002877999,2.591604e-7,0.001113141,0.00001742926,0.003959965,0.755452,0.003753403,0.2337622],"study_design_scores_gemma":[0.0007860092,0.0009258373,0.004839174,0.00153003,0.0001445889,0.000194707,0.003393804,0.06209077,0.03043041,0.1107761,0.7846013,0.000287291],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03975097,0.008543022,0.7753453,0.1581432,0.002124047,0.003001853,0.0001163035,0.001295384,0.01167994],"genre_scores_gemma":[0.9497894,0.002261534,0.04613283,0.00168364,0.0000311767,0.0000445354,0.000003481509,0.000005540838,0.0000478477],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9100384,"threshold_uncertainty_score":0.9927054,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009216745243985884,"score_gpt":0.2873874469861941,"score_spread":0.2781707017422082,"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."}}