{"id":"W2141870796","doi":"10.5539/cis.v6n3p89","title":"Clustering of Web Search Results Based on Document Segmentation","year":2013,"lang":"en","type":"article","venue":"Computer and Information Science","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Cluster analysis; Computer science; Document clustering; Information retrieval; Data mining; Similarity (geometry); Brown clustering; Segmentation; Fuzzy clustering; Correlation clustering; Search engine; CURE data clustering algorithm; Pattern recognition (psychology); Artificial intelligence; Image (mathematics)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007155558,0.00006051101,0.0000738374,0.0003825566,0.0001437483,0.0005563314,0.0005017573,0.0000135391,0.000006181738],"category_scores_gemma":[0.00002198894,0.00004870856,0.0000172898,0.0007133571,0.00009705692,0.007391937,0.0002409105,0.00004300604,0.00005324294],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002353533,"about_ca_system_score_gemma":0.00008107613,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002832506,"about_ca_topic_score_gemma":2.918829e-7,"domain_scores_codex":[0.9989523,0.00002047205,0.0002757056,0.0001514167,0.0004597406,0.000140374],"domain_scores_gemma":[0.9992758,0.00005419131,0.00009541704,0.000309214,0.0001903409,0.0000750276],"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":[0.000008932626,0.00002931236,0.0004165845,0.00004770455,0.000005304095,3.015337e-7,0.003588502,0.06817213,0.001572659,0.004180457,0.00145745,0.9205207],"study_design_scores_gemma":[0.0002836992,0.0000921998,0.006373748,0.00002399218,8.277925e-7,0.000001070298,0.00004378856,0.9907683,0.002040714,0.00001942637,0.0002940133,0.00005821823],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03877427,0.000001717491,0.9577125,0.0005960289,0.0001178089,0.00009348807,0.000003475324,0.00003453637,0.002666173],"genre_scores_gemma":[0.8691401,0.000005469382,0.1302568,0.0005652449,0.00001183238,0.000003881527,0.00000741116,6.96697e-7,0.000008535436],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9225962,"threshold_uncertainty_score":0.5364716,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01516549125906185,"score_gpt":0.2615319963349998,"score_spread":0.246366505075938,"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."}}