{"id":"W3034396897","doi":"10.1007/978-3-030-50578-3_3","title":"Structural Profiling of Web Sites in the Wild","year":2020,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université du Québec à Chicoutimi","funders":"","keywords":"Computer science; Profiling (computer programming); Information retrieval; World Wide Web; Point (geometry); Data mining; Data science; 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":[0.0005495821,0.0002562228,0.0004001266,0.004223466,0.0005653374,0.002095892,0.0006480443,0.0005532101,0.001391825],"category_scores_gemma":[0.00472836,0.0003494096,0.0003737135,0.005596441,0.0003610711,0.002671125,0.0006676847,0.0007050497,0.001567493],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004199297,"about_ca_system_score_gemma":0.000473892,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002976638,"about_ca_topic_score_gemma":0.01081784,"domain_scores_codex":[0.9988649,0.0001935021,0.00008463183,0.0002811371,0.000440827,0.0001350286],"domain_scores_gemma":[0.9954159,0.001745163,0.000622978,0.0008768297,0.000984882,0.0003541327],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004976911,0.000511531,0.4663682,0.0006586461,0.0002461834,0.0008606538,0.002097892,0.01982095,0.06779081,0.02544405,0.02086301,0.3948404],"study_design_scores_gemma":[0.00002464229,0.0002796226,0.4083389,0.0002577511,0.0001824688,0.002936818,0.003712686,0.4389445,0.03568158,0.06378625,0.04575218,0.0001026359],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8727206,0.002430211,0.09701876,0.0007499785,0.0001395372,0.0001398734,0.0113856,0.002409016,0.01300647],"genre_scores_gemma":[0.9370651,0.0008435221,0.04586415,0.00009304858,0.00008605111,0.00006371789,0.0106461,0.0003557893,0.004982413],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004223466,"threshold_uncertainty_score":0.005918622,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02540800676925052,"score_gpt":0.2492081246127359,"score_spread":0.2238001178434853,"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."}}