{"id":"W2941931605","doi":"","title":"Duplicate Removal for Overlapping Clusters: A Study Using Social Media Data.","year":2019,"lang":"en","type":"article","venue":"AAAI Spring Symposium Combining Machine Learning with Knowledge Engineering","topic":"Internet Traffic Analysis and Secure E-voting","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Guelph","funders":"","keywords":"Social media; Computer science; World Wide Web","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":[0.01523422,0.0004977157,0.0009864635,0.00646353,0.003900245,0.002267387,0.002654796,0.001455441,0.0009761848],"category_scores_gemma":[0.09498867,0.0004511838,0.0009763342,0.008072901,0.001232124,0.003473414,0.001280728,0.0007861422,0.0004909681],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001723674,"about_ca_system_score_gemma":0.002385959,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01224087,"about_ca_topic_score_gemma":0.01097403,"domain_scores_codex":[0.9866462,0.00588267,0.0009401601,0.001795709,0.004156074,0.0005791302],"domain_scores_gemma":[0.8708248,0.09710134,0.005081946,0.01339292,0.01277311,0.0008258963],"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":[0.002947906,0.001919387,0.3825569,0.002361226,0.00232116,0.003013985,0.009911004,0.03102955,0.01015815,0.02128586,0.0318065,0.5006884],"study_design_scores_gemma":[0.0003893438,0.001060894,0.2688255,0.0005909428,0.002995062,0.008717704,0.01663842,0.5453398,0.0505934,0.04259067,0.06188737,0.0003709664],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9533845,0.002687626,0.03562893,0.0007547272,0.0002240044,0.0005304944,0.002102224,0.0004493835,0.004238076],"genre_scores_gemma":[0.9544361,0.0005191467,0.03977659,0.0001153393,0.0001204747,0.0002036987,0.002672744,0.0001062674,0.002049604],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01523422,"threshold_uncertainty_score":0.08056718,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01923810525571718,"score_gpt":0.249872969439877,"score_spread":0.2306348641841598,"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."}}