{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002150237,0.0001550402,0.0002029853,0.0001793195,0.0001539919,0.000172362,0.003048449,0.00016231,0.0002003689],"category_scores_gemma":[0.00002252667,0.0001478053,0.00005740082,0.0003606394,0.00001126975,0.0006612892,0.0002627692,0.00020302,0.0001228063],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002640494,"about_ca_system_score_gemma":0.00016822,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001407535,"about_ca_topic_score_gemma":0.001872497,"domain_scores_codex":[0.998718,0.00007223687,0.0001317146,0.000675622,0.0002566122,0.0001458449],"domain_scores_gemma":[0.9975333,0.00004788722,0.0001549708,0.002132794,0.00005850528,0.0000724968],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00004863674,0.0001119159,0.0002497878,0.0001859799,0.0006784981,0.0001952809,0.0007949646,0.00004157907,0.002753345,0.1963086,0.5156013,0.28303],"study_design_scores_gemma":[0.0003544476,0.00002544586,0.002229648,0.00005073073,0.0001418129,0.00001163009,0.001385055,0.04167587,0.0001939507,0.001128845,0.9522446,0.0005580086],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01854354,0.0001575307,0.6885032,0.0008746272,0.003504082,0.0001850781,0.0004361133,0.0008913064,0.2869045],"genre_scores_gemma":[0.2915676,0.0001782957,0.379477,0.0001990764,0.0008274437,0.000001051259,0.04000622,0.0000455689,0.2876977],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4366432,"threshold_uncertainty_score":0.6027323,"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."}}