{"id":"W2113958117","doi":"10.3115/1220175.1220277","title":"Names and similarities on the web","year":2006,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":83,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Ranking (information retrieval); Information retrieval; Quality (philosophy); Scale (ratio); Natural language processing; Artificial intelligence; World Wide Web; Geography","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":[0.001076267,0.0002382369,0.0003724908,0.009661334,0.00140139,0.003834017,0.0004525287,0.0010443,0.00636298],"category_scores_gemma":[0.01161298,0.0002742496,0.0002923632,0.01279767,0.001645023,0.0126765,0.002630759,0.0005244249,0.001748421],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008042553,"about_ca_system_score_gemma":0.0004154407,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002231237,"about_ca_topic_score_gemma":0.002504941,"domain_scores_codex":[0.9977193,0.0006324985,0.0001886223,0.0003788472,0.0009976712,0.00008305134],"domain_scores_gemma":[0.9938146,0.00335241,0.0008415243,0.0008644547,0.0009434499,0.0001836006],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003005863,0.00006679379,0.02566225,0.0005779465,0.0001120038,0.001185534,0.00351049,0.006685924,0.006027403,0.4850241,0.02309674,0.4477503],"study_design_scores_gemma":[0.00003225635,0.00007007433,0.02643469,0.0002662312,0.0001072494,0.002954624,0.003041272,0.02895865,0.005286017,0.6739724,0.2587883,0.00008831352],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4901693,0.03044049,0.2596105,0.008462283,0.0009106358,0.0001942072,0.009682789,0.00387562,0.1966542],"genre_scores_gemma":[0.8877537,0.006725197,0.08033366,0.0005864557,0.0006214323,0.000101232,0.004777904,0.00037212,0.01872834],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009661334,"threshold_uncertainty_score":0.02128625,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007681838690379605,"score_gpt":0.2216454153643378,"score_spread":0.2139635766739582,"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."}}