{"id":"W2111094784","doi":"10.1109/iccit.2008.210","title":"Social Network Analysis on Name Disambiguation and More","year":2008,"lang":"en","type":"article","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Disk formatting; String metric; Social network analysis; Information retrieval; Similarity (geometry); String (physics); Correct name; Social network (sociolinguistics); Natural language processing; Graph; World Wide Web; Artificial intelligence; String searching algorithm; Social media; Theoretical computer science; Pattern matching; Mathematics","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":[0.01197652,0.001146366,0.001325002,0.01703429,0.002391767,0.0051069,0.001862599,0.002105986,0.003941169],"category_scores_gemma":[0.06350293,0.0004216034,0.001012874,0.01415764,0.002929009,0.01326179,0.003076563,0.001044432,0.0007560844],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002627027,"about_ca_system_score_gemma":0.001495281,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005515228,"about_ca_topic_score_gemma":0.004927299,"domain_scores_codex":[0.9826003,0.009672861,0.00107414,0.00276464,0.003554305,0.0003337928],"domain_scores_gemma":[0.9338813,0.04440994,0.008522058,0.007671461,0.00487535,0.0006397531],"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":[0.0006521411,0.0003876919,0.07885966,0.001364599,0.0006530103,0.0007214586,0.002813305,0.09989691,0.005575793,0.1715251,0.01629448,0.6212559],"study_design_scores_gemma":[0.00006256446,0.0001585565,0.0300368,0.0002957965,0.0002442264,0.0007220625,0.002230343,0.6487479,0.009564059,0.2811919,0.02657237,0.0001733677],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2391409,0.00688582,0.7208585,0.009169876,0.0004610835,0.0003952919,0.003121354,0.001406414,0.01856077],"genre_scores_gemma":[0.7710302,0.002032682,0.220265,0.0005068746,0.0004991405,0.0001971229,0.002216911,0.0001910913,0.003060982],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01703429,"threshold_uncertainty_score":0.0633387,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2179606429412178,"score_gpt":0.4426846600958866,"score_spread":0.2247240171546688,"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."}}