{"id":"W2047032662","doi":"10.3115/1072228.1072372","title":"Concept discovery from text","year":2002,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":185,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"WordNet; Cluster analysis; Computer science; Set (abstract data type); Centroid; Cluster (spacecraft); Task (project management); Similarity (geometry); Artificial intelligence; Data mining; Feature vector; Information retrieval; Quality (philosophy); Natural language processing; Image (mathematics)","routes":{"ca_aff":true,"ca_fund":true,"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":[0.003287466,0.00165189,0.001721208,0.01300771,0.001800321,0.003542377,0.003028785,0.001726685,0.005754899],"category_scores_gemma":[0.02128439,0.0006149966,0.001793809,0.009066447,0.001120659,0.006504131,0.003505487,0.001773284,0.003392384],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001339904,"about_ca_system_score_gemma":0.002318621,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002191297,"about_ca_topic_score_gemma":0.003067634,"domain_scores_codex":[0.9941982,0.001327005,0.0004219269,0.001570582,0.002247813,0.0002344813],"domain_scores_gemma":[0.9889746,0.007367627,0.0005880472,0.001011461,0.001729138,0.0003290827],"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":[0.0005401606,0.0004128398,0.006377529,0.002168478,0.0002508136,0.001558125,0.001939727,0.0104662,0.012993,0.03361594,0.03156692,0.8981103],"study_design_scores_gemma":[0.0002549499,0.0004071789,0.007760765,0.0008696645,0.0004602295,0.00364187,0.004078102,0.4170431,0.05412566,0.3012054,0.2099089,0.0002440606],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07344726,0.004169924,0.8865746,0.002543055,0.0004378755,0.001586944,0.01429671,0.005854557,0.01108905],"genre_scores_gemma":[0.1155615,0.001600348,0.8588792,0.0003973458,0.0003033576,0.001039292,0.01839267,0.0002734238,0.003552773],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01300771,"threshold_uncertainty_score":0.019252,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01340747817539674,"score_gpt":0.2381662981531921,"score_spread":0.2247588199777953,"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."}}