{"id":"W2025635041","doi":"10.1145/1754393.1754394","title":"Detecting visually similar Web pages","year":2010,"lang":"en","type":"article","venue":"ACM Transactions on Internet Technology","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":98,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Gestalt psychology; Web page; Similarity (geometry); World Wide Web; Realization (probability); Information retrieval; Phishing; Artificial intelligence; The Internet; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005992999,0.0004888182,0.0008055288,0.009164144,0.0007460538,0.002415271,0.0009287476,0.00146083,0.00196959],"category_scores_gemma":[0.00633232,0.0003846395,0.0006626892,0.002426852,0.0008136267,0.003225347,0.002116546,0.0007646814,0.0008349479],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005783563,"about_ca_system_score_gemma":0.0005241713,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00131258,"about_ca_topic_score_gemma":0.001652736,"domain_scores_codex":[0.9987522,0.0002010476,0.00007218702,0.0002504573,0.0006057977,0.0001183126],"domain_scores_gemma":[0.9969558,0.0008929883,0.0006645949,0.0004707778,0.0008219709,0.0001939281],"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":[0.0006607313,0.0006925837,0.04632932,0.0006917213,0.000393151,0.001465356,0.001651998,0.02184879,0.1114804,0.04844881,0.007202583,0.7591345],"study_design_scores_gemma":[0.00009035959,0.000621057,0.06597039,0.000107342,0.000298376,0.006951423,0.001430303,0.7313923,0.0726891,0.100178,0.02005497,0.0002162759],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3022366,0.001018787,0.6848692,0.0004253708,0.00009900704,0.000352475,0.0004129427,0.002450381,0.008135105],"genre_scores_gemma":[0.742507,0.0002844222,0.2543097,0.0001274315,0.000113351,0.0001281765,0.0005013846,0.0001161749,0.001912124],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009164144,"threshold_uncertainty_score":0.006588876,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0128471117380512,"score_gpt":0.2488890762803148,"score_spread":0.2360419645422636,"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."}}