{"id":"W2808279896","doi":"10.24963/ijcai.2018/411","title":"Cost-Effective Active Learning for Hierarchical Multi-Label Classification","year":2018,"lang":"en","type":"article","venue":"","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":41,"is_retracted":false,"has_abstract":true,"ca_institutions":"Novelis (Canada)","funders":"National Natural Science Foundation of China","keywords":"Computer science; Machine learning; Exploit; Selection (genetic algorithm); Artificial intelligence; Active learning (machine learning); Annotation; Multi-label classification; Hierarchy; Labeled data","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":[0.0003132019,0.0001139132,0.0001182829,0.00007944719,0.0003503335,0.0001066072,0.0003738355,0.00006466813,0.00001910918],"category_scores_gemma":[0.0003277041,0.00009429195,0.00004584024,0.0002201702,0.0000840647,0.0002473566,0.0001096823,0.0002799805,0.0001323042],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004419939,"about_ca_system_score_gemma":0.00003406075,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002138991,"about_ca_topic_score_gemma":0.0000075489,"domain_scores_codex":[0.9989185,0.0001467881,0.000120339,0.0004150149,0.0001370981,0.0002622442],"domain_scores_gemma":[0.9990602,0.0003764294,0.00007193736,0.0002265847,0.000175857,0.00008899754],"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":[0.0000180691,0.00007845985,0.0002580951,0.000003986839,0.00001357774,4.453669e-7,0.0008859846,0.00002266359,0.001290802,0.02087357,0.0002125155,0.9763418],"study_design_scores_gemma":[0.0008407087,0.000417775,0.009920937,0.000007077551,0.000003574479,0.000003551908,0.00005238784,0.969456,0.002377479,0.0006470475,0.01613707,0.0001363605],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004099963,0.000005522754,0.9901153,0.00112594,0.0002824567,0.0007909558,9.382476e-7,0.0003306911,0.003248226],"genre_scores_gemma":[0.5131649,0.000001690226,0.4816507,0.0002949678,0.0003475462,0.0004088508,0.000008216116,0.00001359675,0.004109502],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9762055,"threshold_uncertainty_score":0.3845114,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06644345468817692,"score_gpt":0.3616965684584418,"score_spread":0.2952531137702649,"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."}}