{"id":"W4366850523","doi":"10.1145/3543507.3583457","title":"CEIL: A General Classification-Enhanced Iterative Learning Framework for Text Clustering","year":2023,"lang":"en","type":"article","venue":"","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Cluster analysis; Computer science; Artificial intelligence; Conceptual clustering; Document clustering; Correlation clustering; Canopy clustering algorithm; CURE data clustering algorithm; Brown clustering; Representation (politics); Machine learning; Feature learning; Data mining; Pattern recognition (psychology)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002351068,0.001736213,0.001781994,0.003574763,0.001275952,0.001997995,0.0059751,0.002686099,0.002846171],"category_scores_gemma":[0.00766028,0.0007390607,0.001767161,0.003327578,0.001572307,0.003333452,0.002963496,0.003394929,0.002989473],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002293152,"about_ca_system_score_gemma":0.003041692,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00943477,"about_ca_topic_score_gemma":0.01749754,"domain_scores_codex":[0.9973733,0.0006310235,0.0001490048,0.0008753842,0.000713898,0.0002574766],"domain_scores_gemma":[0.9972052,0.0009298165,0.0002503442,0.0004724898,0.0009836435,0.0001584298],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003601617,0.0003738739,0.002388707,0.0004400902,0.000210932,0.0002499076,0.0007583314,0.3470389,0.01961406,0.03109887,0.01889749,0.5785686],"study_design_scores_gemma":[0.0000195343,0.00005113184,0.0002235977,0.00001515728,0.00001554253,0.00005998974,0.00004448605,0.9773032,0.004068018,0.01433805,0.003833543,0.00002775423],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002561413,0.0002008113,0.9946769,0.00008874716,0.00002778874,0.00009307321,0.0001405838,0.001663715,0.0005468602],"genre_scores_gemma":[0.09370536,0.0003146265,0.8954297,0.0004385472,0.0001300982,0.00070581,0.002553212,0.0006761068,0.006046496],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00943477,"threshold_uncertainty_score":0.01875967,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05283188345815129,"score_gpt":0.3228554519467172,"score_spread":0.2700235684885659,"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."}}