{"id":"W2407195633","doi":"","title":"A Statistical Model for Measuring Structural Similarity between Webpages","year":2015,"lang":"en","type":"article","venue":"Recent Advances in Natural Language Processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Similarity (geometry); Computer science; Web page; Statistical model; Task (project management); Measure (data warehouse); Artificial intelligence; Structural similarity; Information retrieval; Natural language processing; Data mining; Image (mathematics); World Wide Web","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0007316886,0.000352984,0.0004514612,0.0002490874,0.0001764213,0.0003256652,0.001189473,0.0001583949,0.000002336149],"category_scores_gemma":[0.001160381,0.0002966516,0.00005848223,0.0006586929,0.0001007385,0.002891917,0.0003455554,0.000640359,0.000001481232],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003526289,"about_ca_system_score_gemma":0.0002663356,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008345354,"about_ca_topic_score_gemma":0.00005534052,"domain_scores_codex":[0.9973401,0.0000868943,0.0004994692,0.0007464501,0.0006120742,0.0007149944],"domain_scores_gemma":[0.998624,0.0002110172,0.0002386282,0.0003631389,0.0003637637,0.0001994669],"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":[0.00005355318,0.00001944196,0.001011107,0.0002536596,0.000006176904,0.00002752576,0.002385884,0.0006280111,0.0002898454,0.002732604,0.00005419582,0.992538],"study_design_scores_gemma":[0.0009124437,0.00005342215,0.0000819079,0.0002529952,0.0000156672,0.00002360004,0.0002431011,0.9193919,0.004611324,0.07322766,0.0005789079,0.0006071197],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00307002,0.2195458,0.7755873,0.000401941,0.0002379502,0.0004008053,0.00002310989,0.0006076464,0.0001254508],"genre_scores_gemma":[0.5257224,0.0001053799,0.4738537,0.0001090471,0.0001061693,0.0000293722,0.0000264472,0.00001900534,0.00002846334],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9919309,"threshold_uncertainty_score":0.9999486,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03928171059953837,"score_gpt":0.3505640517397068,"score_spread":0.3112823411401684,"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."}}