{"id":"W4255377955","doi":"10.22215/etd/2014-10139","title":"Structured Web Data Extraction: University Domain","year":2014,"lang":"en","type":"dissertation","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"University of Toronto","keywords":"Computer science; World Wide Web; Information retrieval; Data Web; Semantic Web; Web modeling; Information extraction; Unstructured data; Web intelligence; Domain (mathematical analysis); Web page; Web search engine; Web development; Data mining; Big data","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"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":[0.001658891,0.0004586398,0.0005131735,0.006047106,0.001379237,0.002735437,0.0008094191,0.001262871,0.003865214],"category_scores_gemma":[0.007023256,0.0003469095,0.0005551216,0.0114396,0.0007767625,0.004044296,0.002144032,0.001098496,0.004243785],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009040207,"about_ca_system_score_gemma":0.002598995,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003189045,"about_ca_topic_score_gemma":0.005416116,"domain_scores_codex":[0.9978918,0.0005211011,0.0001827275,0.0003360315,0.0009646469,0.0001037134],"domain_scores_gemma":[0.9960076,0.001304704,0.0002847768,0.001039181,0.001168318,0.0001954355],"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":[0.0002910858,0.0006873364,0.01278178,0.001167346,0.00008405768,0.0006437919,0.001146271,0.00546748,0.008391274,0.03979247,0.1077151,0.821832],"study_design_scores_gemma":[0.0001080662,0.0002019963,0.02100181,0.0009574071,0.0001991712,0.001826592,0.002921387,0.1153974,0.09156494,0.1194633,0.646208,0.0001500412],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1580183,0.01337929,0.6749485,0.01300502,0.001008378,0.001562641,0.05797601,0.0182425,0.06185933],"genre_scores_gemma":[0.3612504,0.009840143,0.5245014,0.001488065,0.0008125406,0.0006747022,0.08092029,0.001346789,0.01916557],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006047106,"threshold_uncertainty_score":0.01293039,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01875610286432536,"score_gpt":0.2644575438322607,"score_spread":0.2457014409679353,"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."}}