{"id":"W4223909614","doi":"10.13052/jwe1540-9589.2146","title":"Integrated-Block: A New Combination Model to Improve Web Page Segmentation","year":2022,"lang":"en","type":"article","venue":"Journal of Web Engineering","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; Web page; Segmentation; Block (permutation group theory); Information retrieval; Gestalt psychology; Context (archaeology); Similarity (geometry); Artificial intelligence; Data mining; World Wide Web; Perception; 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":[0.0004198718,0.00008562873,0.0001418371,0.0003847541,0.00006586208,0.0001075181,0.0004990518,0.00001544712,0.000008459789],"category_scores_gemma":[0.00005371206,0.0000832435,0.00007652888,0.0005434383,0.000001811664,0.0004253169,0.0001644577,0.0002390411,0.000003367797],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001613451,"about_ca_system_score_gemma":0.0001992822,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007571942,"about_ca_topic_score_gemma":0.00000103823,"domain_scores_codex":[0.9990988,0.00002021133,0.0002856574,0.0001207044,0.0003414103,0.0001332676],"domain_scores_gemma":[0.9994817,0.00002639127,0.0001394596,0.0001669551,0.00006794777,0.0001175601],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000004024346,0.00002965422,0.00002740009,0.000004967331,0.00003350388,0.00001443085,0.0004283608,0.7363743,0.2509311,0.000536136,0.003731661,0.007884499],"study_design_scores_gemma":[0.000357309,0.0001342705,0.00002286779,0.00001606292,0.00001616059,0.0000370322,0.0001027704,0.9952107,0.002424235,0.00004674126,0.001540379,0.00009144162],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2238487,0.0000600595,0.774681,0.0008714591,0.0003989641,0.00004084518,0.000007511472,0.00004346607,0.0000480305],"genre_scores_gemma":[0.9063005,0.000007925282,0.09329449,0.00009819671,0.0000663007,0.000003072925,0.00000271836,0.000008235421,0.0002185101],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6824518,"threshold_uncertainty_score":0.3394571,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01158827969045619,"score_gpt":0.2223981886257918,"score_spread":0.2108099089353356,"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."}}