{"id":"W3169038444","doi":"10.13052/jwe1540-9589.2042","title":"A New Semantic Approach to Improve Webpage Segmentation","year":2021,"lang":"en","type":"article","venue":"Journal of Web Engineering","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Segmentation; Web page; Computer science; Merge (version control); Information retrieval; Artificial intelligence; Data mining; 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":[],"consensus_categories":[],"category_scores_codex":[0.0002221344,0.00007431267,0.0001574742,0.0001725911,0.00002038229,0.0001464095,0.0003223494,0.00002238386,0.0000049577],"category_scores_gemma":[0.00007904601,0.00006866788,0.00008515189,0.0004799121,0.000001325506,0.000372141,0.00009152287,0.0001247673,0.000009344106],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003536882,"about_ca_system_score_gemma":0.0001309122,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002977249,"about_ca_topic_score_gemma":4.140711e-7,"domain_scores_codex":[0.9992747,0.00001274303,0.0002404714,0.0001255627,0.0002142371,0.0001322932],"domain_scores_gemma":[0.9994658,0.00002608815,0.00008399619,0.0002037053,0.00007557398,0.0001448323],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000004029665,0.00008601732,0.0002282078,0.00008053522,0.0002423117,0.0002061661,0.001151805,0.1408862,0.7840027,0.002036408,0.008421392,0.06265427],"study_design_scores_gemma":[0.0007367987,0.0001078235,0.0008957228,0.0001344041,0.00008008502,0.0005199006,0.0001899739,0.9313874,0.05807224,0.00004518241,0.007538662,0.0002918587],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02409111,0.0002032852,0.9747698,0.0003518003,0.0003179903,0.00001680806,7.998019e-7,0.00002247641,0.0002259176],"genre_scores_gemma":[0.3787011,0.00002033623,0.62063,0.00007318219,0.0002578009,6.214097e-7,0.000001100668,0.000007245809,0.000308608],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7905012,"threshold_uncertainty_score":0.2800195,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008387315172901347,"score_gpt":0.2116404610791718,"score_spread":0.2032531459062705,"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."}}