{"id":"W4395669106","doi":"10.18280/ijsse.140214","title":"Performance Analysis of Regex-Based Processing for Dark Web Targeted Crawling","year":2024,"lang":"en","type":"article","venue":"International Journal of Safety and Security Engineering","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Universitas Telkom","keywords":"Crawling; Computer science; Medicine","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005516219,0.00008170938,0.000218271,0.0007479445,0.00003421912,0.0001358765,0.0004090171,0.00003175253,0.000003277094],"category_scores_gemma":[0.00008145801,0.00007291387,0.0001688703,0.0005717892,0.00001448782,0.000557552,0.00004439516,0.000126621,1.537658e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003589885,"about_ca_system_score_gemma":0.00009379902,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002507373,"about_ca_topic_score_gemma":0.000001446369,"domain_scores_codex":[0.9990777,0.000008657355,0.0004011472,0.0001292979,0.000286499,0.00009667822],"domain_scores_gemma":[0.9993114,0.000147581,0.0001371815,0.00008176327,0.0002754569,0.00004663846],"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":[0.0001970495,0.0001162904,0.003072045,0.0007791321,0.005384906,0.00009214164,0.003727562,0.839747,0.02411079,0.005708268,0.000107438,0.1169573],"study_design_scores_gemma":[0.0001764274,0.00003327279,0.0006932893,0.000314784,0.0001862888,0.00001408878,0.00001988545,0.9950275,0.001877283,0.00003845578,0.001542826,0.00007591814],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1694036,0.001225753,0.8283328,0.0005844324,0.0003349158,0.00002115405,0.0000378316,0.00002884213,0.00003061041],"genre_scores_gemma":[0.980444,0.0001547589,0.0192586,0.00002107387,0.0000977543,7.011858e-7,0.00001282601,0.000004920961,0.000005321557],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8110404,"threshold_uncertainty_score":0.2973341,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006726398278615291,"score_gpt":0.2372830568169749,"score_spread":0.2305566585383596,"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."}}