{"id":"W4405138524","doi":"10.1007/s00521-024-10408-7","title":"AI-UNet: Attention Information-based deep URL Network for adult webpage classification","year":2024,"lang":"en","type":"article","venue":"Neural Computing and Applications","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"Science and Engineering Research Board","keywords":"Computational Science and Engineering; Computer science; Artificial intelligence; Web page; Information retrieval; Machine learning; 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":[0.0002085353,0.0001093573,0.000104239,0.0001078352,0.0004443607,0.000605012,0.0002972314,0.00004955917,0.000001412543],"category_scores_gemma":[0.00002440986,0.0001033481,0.00007309163,0.000533075,0.00003137456,0.0004511162,0.00005861632,0.0001204769,0.00003489169],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001813879,"about_ca_system_score_gemma":0.00003305908,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001182564,"about_ca_topic_score_gemma":0.000004544048,"domain_scores_codex":[0.9990811,0.00002764168,0.0002727055,0.0003039245,0.0001184484,0.0001962284],"domain_scores_gemma":[0.9991367,0.0002264264,0.00008135612,0.0003255768,0.0001657633,0.00006417515],"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":[0.000002815315,0.00002871625,0.0006924612,0.0001370318,0.00003030272,3.369026e-7,0.0001609384,0.0142189,0.0002906014,0.2548395,0.009963576,0.7196349],"study_design_scores_gemma":[0.0001069357,0.00001970305,0.001149122,0.00003046659,0.00002373104,0.000002967029,0.00002320665,0.9496553,0.00001526625,0.001527254,0.04733412,0.0001119119],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002807896,0.0001410157,0.9901798,0.005872045,0.0001282723,0.0002545041,0.00002423391,0.0004021116,0.0001901074],"genre_scores_gemma":[0.9548055,0.000008061031,0.04251398,0.001727143,0.0003317433,0.0001325109,0.0004314036,0.000007908719,0.00004173595],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9519976,"threshold_uncertainty_score":0.5834144,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01393270084253853,"score_gpt":0.2715575898749068,"score_spread":0.2576248890323682,"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."}}