{"id":"W1485672862","doi":"","title":"BlogScope: a system for online analysis of high volume text streams","year":2007,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":79,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Blogosphere; Computer science; Ranking (information retrieval); Relevance (law); Information retrieval; Volume (thermodynamics); Set (abstract data type); Data stream mining; World Wide Web; Data mining; Data science; The Internet","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005910666,0.0001030227,0.0003933531,0.0005150837,0.00005257971,0.00005133733,0.0007319278,0.00004913893,0.00002888043],"category_scores_gemma":[0.00003715049,0.00008192335,0.0002355797,0.002133127,0.00002481172,0.000164167,0.0001296768,0.0000387949,0.00001226607],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003085721,"about_ca_system_score_gemma":0.00003150712,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001079652,"about_ca_topic_score_gemma":0.0007277448,"domain_scores_codex":[0.9987894,0.00001981785,0.0003916688,0.0003477498,0.0002144251,0.0002369013],"domain_scores_gemma":[0.9987252,0.0001465629,0.0001539763,0.0007472652,0.0001444642,0.00008257186],"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":[0.00006391127,0.001265972,0.02967839,0.0002935708,0.01055621,0.0000496018,0.0007387627,0.01061249,0.002756321,0.1688699,0.008242382,0.7668725],"study_design_scores_gemma":[0.0004688052,0.0001538637,0.01768842,0.00003302663,0.001552384,0.000002953353,0.000494863,0.9726841,0.003337085,0.00004892947,0.003274124,0.0002614832],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1587836,0.00003124011,0.8401632,0.0001380237,0.00005545859,0.00005397616,0.0001083711,0.0001001413,0.0005659792],"genre_scores_gemma":[0.8598476,0.000001607432,0.1383023,0.00004568805,0.00003074405,0.000002334709,0.0001156252,0.000004136717,0.001649951],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9620716,"threshold_uncertainty_score":0.3340738,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01347909899140815,"score_gpt":0.2599205676831217,"score_spread":0.2464414686917135,"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."}}