{"id":"W4240635190","doi":"10.1002/(sici)1097-4571(2000)51:9<793::aid-asi20>3.0.co;2-q","title":"Rating news documents for similarity","year":2000,"lang":"en","type":"article","venue":"Journal of the American Society for Information Science","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Acadia University; Dalhousie University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Personalization; Computer science; Information retrieval; Feature (linguistics); Similarity (geometry); Precision and recall; Event (particle physics); Representation (politics); World Wide Web; News aggregator; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":[],"domain":null,"study_design":"simulation_or_modeling","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"bench_or_experimental","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"}],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001793387,0.00006782472,0.0001538024,0.00006029557,0.0006422549,0.0005301123,0.001876359,0.00001156343,0.000003689924],"category_scores_gemma":[0.0002685389,0.00004339544,0.0004473704,0.001289221,0.000303477,0.005978408,0.000105151,0.00008927078,0.000003465154],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008209098,"about_ca_system_score_gemma":0.0002502213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001621191,"about_ca_topic_score_gemma":4.779849e-7,"domain_scores_codex":[0.998705,0.00001324337,0.0004309457,0.00008473395,0.0005565949,0.0002094845],"domain_scores_gemma":[0.9981651,0.0001153461,0.0009295368,0.0003049668,0.0004099724,0.00007512783],"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":[0.00002675336,0.00003544282,0.0006155176,0.00001934134,0.00007202191,6.036498e-8,0.006048128,0.005453134,0.001204698,0.002260956,0.04585082,0.9384131],"study_design_scores_gemma":[0.001306664,0.0004873668,0.003615485,0.0000636583,0.00008347639,0.00005856464,0.004968404,0.5316528,0.00319974,0.002146953,0.452056,0.0003609323],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1783186,0.00001247627,0.8141091,0.006269248,0.0002921745,0.0002058282,0.00002158659,0.00002423024,0.0007467189],"genre_scores_gemma":[0.5565748,0.00005595871,0.4341639,0.00887183,0.0001303299,0.000007396231,0.000001971015,0.000003664593,0.0001901973],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9380522,"threshold_uncertainty_score":0.5111885,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01678433477910057,"score_gpt":0.3004383908694803,"score_spread":0.2836540560903797,"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."}}