{"id":"W2059984218","doi":"10.1145/584792.584802","title":"Topic-oriented collaborative crawling","year":2002,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Web crawler; Computer science; Crawling; Focused crawler; Hash function; Information retrieval; Web page; World Wide Web; Classifier (UML); Data mining; Static web page; Web navigation; 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":[0.00006484792,0.00005323449,0.00007472149,0.00005433814,0.00008222267,0.00009641401,0.0003188785,0.00001845466,0.0002680962],"category_scores_gemma":[0.00003616495,0.00004314092,0.00002563542,0.0006487192,0.00001262241,0.0002966766,0.0000950672,0.00004058221,0.000272461],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000009048543,"about_ca_system_score_gemma":0.000006910721,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001475164,"about_ca_topic_score_gemma":0.00000701133,"domain_scores_codex":[0.9994498,0.0000222483,0.00009202562,0.0001949736,0.0001192622,0.0001216583],"domain_scores_gemma":[0.9995038,0.00003194667,0.00002477531,0.0003298979,0.00006331519,0.0000462785],"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":[0.000002485776,0.0003565603,0.00751557,0.00001329995,0.0002023456,0.0001151706,0.009997931,0.0001609104,0.003977247,0.4631759,0.1274627,0.3870198],"study_design_scores_gemma":[0.0004737306,0.0001007372,0.0006919188,0.00001716877,0.0000164973,0.000007736453,0.0005505109,0.6573646,0.01073367,0.00067409,0.328978,0.0003913224],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01078986,0.00012852,0.9127024,0.002579903,0.0001745846,0.00003349998,0.000002433906,0.0002350087,0.07335383],"genre_scores_gemma":[0.8569707,0.00002355447,0.1253981,0.0006452831,0.00007121033,0.000004143693,0.000002746121,0.000003175551,0.0168811],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8461809,"threshold_uncertainty_score":0.3502026,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01527913115665228,"score_gpt":0.2217458562176939,"score_spread":0.2064667250610417,"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."}}