{"id":"W1989265129","doi":"10.1108/17440081011090220","title":"Topic‐based web site summarization","year":2010,"lang":"en","type":"article","venue":"International Journal of Web Information Systems","topic":"Advanced Text Analysis Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Automatic summarization; Computer science; Multi-document summarization; Information retrieval; Cluster analysis; Web page; Set (abstract data type); Web modeling; World Wide Web; Artificial intelligence","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":[0.0006305644,0.0001000865,0.0001606998,0.0006900312,0.00004532178,0.000515765,0.001079472,0.00007766631,0.00002002521],"category_scores_gemma":[0.0001893156,0.00008652479,0.0001168076,0.0002575776,0.00002222876,0.0047418,0.00007832653,0.0002479391,0.00006771509],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008702617,"about_ca_system_score_gemma":0.0002169317,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009963441,"about_ca_topic_score_gemma":0.000009080598,"domain_scores_codex":[0.9980329,0.00003779696,0.0008696113,0.00007105814,0.0008834262,0.0001052164],"domain_scores_gemma":[0.996726,0.00007225327,0.001082905,0.0002399991,0.001806715,0.00007209402],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001367612,0.0003191156,0.0480145,0.0001140314,0.0006047512,0.00009136744,0.002780234,0.03554341,0.1251068,0.5873153,0.02472283,0.1752509],"study_design_scores_gemma":[0.0009234048,0.00006501008,0.001203311,0.00008253694,0.00001174833,0.000219954,0.00005096572,0.5631406,0.005108007,0.0005761393,0.4284135,0.0002048174],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02785237,0.00001726739,0.9628506,0.0009637626,0.003649471,0.00009805126,0.000008973365,0.0001034771,0.004456084],"genre_scores_gemma":[0.9735259,0.00001149054,0.02566689,0.0003397287,0.0003564776,0.000004712815,0.00001618291,0.000003832481,0.00007477289],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9456735,"threshold_uncertainty_score":0.4973534,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006202646875641511,"score_gpt":0.260992364161183,"score_spread":0.2547897172855415,"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."}}