{"id":"W2752935397","doi":"10.1504/ijwet.2017.086415","title":"Estimating similarity of rich internet pages using visual information","year":2017,"lang":"en","type":"article","venue":"International Journal of Web Engineering and Technology","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Similarity (geometry); The Internet; Information retrieval; World Wide Web; Artificial intelligence; Image (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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002482849,0.00006284747,0.0001414212,0.0005247704,0.00003664367,0.0001548176,0.0008730736,0.0000588165,0.000001328266],"category_scores_gemma":[0.0005059193,0.00005742209,0.0000315743,0.00007771666,0.0000413884,0.0009159083,0.0003053742,0.0001565248,5.898546e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002061329,"about_ca_system_score_gemma":0.00003346743,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001153083,"about_ca_topic_score_gemma":8.927821e-7,"domain_scores_codex":[0.9993848,0.000005014404,0.0003011248,0.0000607247,0.0001735092,0.00007489068],"domain_scores_gemma":[0.9990516,0.00002992151,0.0004609448,0.0001493234,0.0002845255,0.00002372297],"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":[0.00004301131,0.0002523208,0.109729,0.0001871377,0.001912118,0.0002775631,0.001383051,0.07417066,0.09045382,0.05852283,0.000834253,0.6622342],"study_design_scores_gemma":[0.000238717,0.0000485017,0.001121412,0.0001227535,0.00001305044,0.0002020073,0.0000239294,0.991363,0.006008041,0.0003232119,0.0004739512,0.00006145942],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5692784,0.00004606482,0.4294312,0.0007888617,0.0003956385,0.00000795442,0.000002134031,0.00002153634,0.0000282822],"genre_scores_gemma":[0.8970328,0.00001676682,0.1028844,0.00000798379,0.00005233412,2.206727e-7,7.282356e-7,0.000002032348,0.000002712058],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9171923,"threshold_uncertainty_score":0.2341605,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01111159971823721,"score_gpt":0.2672239810553102,"score_spread":0.256112381337073,"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."}}