{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002399356,0.0007408455,0.001338182,0.01653385,0.0006936474,0.002606048,0.000754517,0.001066504,0.006618119],"category_scores_gemma":[0.02622815,0.0002266584,0.0008792736,0.01084077,0.0003660322,0.002836622,0.0009803071,0.000769039,0.00506704],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006452097,"about_ca_system_score_gemma":0.0006877679,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002244491,"about_ca_topic_score_gemma":0.00398213,"domain_scores_codex":[0.993893,0.000927205,0.0008908021,0.0008169386,0.003198722,0.0002732521],"domain_scores_gemma":[0.987592,0.004014936,0.001651832,0.001126805,0.005113652,0.0005007175],"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":[0.00139354,0.000660562,0.1432793,0.0009147043,0.0007215476,0.0004138865,0.0009163301,0.004056543,0.0271406,0.002589229,0.02735192,0.7905618],"study_design_scores_gemma":[0.0004034733,0.004300661,0.635538,0.0003190014,0.001205076,0.004881992,0.003968736,0.2250625,0.05032617,0.01162069,0.06190478,0.0004687957],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7745066,0.008060239,0.1573885,0.001093996,0.001009625,0.001989167,0.01084002,0.003750401,0.04136153],"genre_scores_gemma":[0.8858247,0.001614892,0.09575031,0.0001013352,0.0007779512,0.0003246063,0.008982409,0.0001666336,0.006457236],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01653385,"threshold_uncertainty_score":0.02213979,"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."}}