{"id":"W4252745723","doi":"10.32920/ryerson.14668263.v1","title":"Researchgate.net crawler and a new contribution determines sequence (CDS) method","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Web crawler; Crawling; Scripting language; Computer science; Focused crawler; World Wide Web; Field (mathematics); Java; Information retrieval; Data mining; Data science; The Internet; Web server; Static web page; Operating system; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.00385447,0.001530175,0.001320596,0.0143479,0.001018152,0.00302155,0.001427238,0.001404122,0.008224028],"category_scores_gemma":[0.02053499,0.001024258,0.00133629,0.008686018,0.0006155015,0.003248312,0.001927129,0.00113548,0.005521446],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001391957,"about_ca_system_score_gemma":0.004028544,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008970808,"about_ca_topic_score_gemma":0.01318391,"domain_scores_codex":[0.9952154,0.0009335162,0.0007096022,0.00120353,0.001768139,0.0001697454],"domain_scores_gemma":[0.9920763,0.003228568,0.0004681474,0.001520941,0.00237523,0.0003308646],"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":[0.0005574642,0.0003113488,0.01314041,0.00177803,0.0004075202,0.000438692,0.0008898772,0.009458526,0.009064628,0.03106831,0.1053004,0.8275847],"study_design_scores_gemma":[0.0007200508,0.0003422456,0.01410954,0.0003268965,0.0003974191,0.001680956,0.000490335,0.4193676,0.03220912,0.03923316,0.4908439,0.0002788002],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02254741,0.001116621,0.8585488,0.000742081,0.0005572446,0.001701369,0.01750779,0.08375985,0.0135189],"genre_scores_gemma":[0.04776071,0.000400123,0.9154879,0.0001216933,0.0001360558,0.001061737,0.01716909,0.005586522,0.01227613],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0143479,"threshold_uncertainty_score":0.02751213,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09491528026853427,"score_gpt":0.3891979685653524,"score_spread":0.2942826882968181,"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."}}