{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001118277,0.001452734,0.001322489,0.007802607,0.0007351345,0.001642566,0.001290589,0.001411903,0.002167472],"category_scores_gemma":[0.002546899,0.0004915084,0.002203332,0.005428493,0.00115251,0.004071897,0.001515801,0.001147558,0.001666694],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009200326,"about_ca_system_score_gemma":0.0015901,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005086076,"about_ca_topic_score_gemma":0.006864782,"domain_scores_codex":[0.9983785,0.0001940771,0.000138696,0.0004897338,0.0006634122,0.0001355658],"domain_scores_gemma":[0.9987067,0.0002180685,0.0001506822,0.0002250815,0.00062624,0.00007321005],"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":[0.0004487941,0.0004965721,0.006575519,0.0007410759,0.0002855422,0.0002348094,0.0005039462,0.05457326,0.07436875,0.01766409,0.01062992,0.8334777],"study_design_scores_gemma":[0.00005015636,0.0001943622,0.006353241,0.00007653213,0.0002105886,0.0004268135,0.0002570614,0.9048557,0.04125565,0.02398377,0.02225836,0.00007772066],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01623477,0.0006612304,0.9750063,0.0001321543,0.00009411589,0.0001606906,0.0003966056,0.004751304,0.002562907],"genre_scores_gemma":[0.1922137,0.0006402297,0.8010054,0.0002500134,0.000156947,0.0002498513,0.001982754,0.0007553951,0.002745625],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007802607,"threshold_uncertainty_score":0.01011294,"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."}}