{"id":"W4398866345","doi":"10.7910/dvn/qtoiqf/dxetjb","title":"scrapers_newspapers.tab","year":2019,"lang":"en","type":"dataset","venue":"Harvard Dataverse","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Newspaper; Scraper site; Geography; Computer science; World Wide Web; Advertising; Business","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0008691587,0.003414461,0.001742279,0.007485982,0.001119828,0.003908861,0.003586013,0.002881237,0.1887491],"category_scores_gemma":[0.005744179,0.001096044,0.001535692,0.01105068,0.0007172464,0.002245046,0.00242656,0.002167737,0.2618485],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001998293,"about_ca_system_score_gemma":0.002182093,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02955721,"about_ca_topic_score_gemma":0.04597117,"domain_scores_codex":[0.9989471,0.0001709498,0.0001134386,0.0003050345,0.0002307385,0.0002327122],"domain_scores_gemma":[0.9973118,0.0007564963,0.0002667682,0.0006858552,0.0006508438,0.0003283171],"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":[0.0000309797,0.00001223301,0.0002296113,0.0004224918,0.00001634767,0.00001119478,0.00001379504,0.00009353596,0.00006727973,0.0002616707,0.9978356,0.001005408],"study_design_scores_gemma":[0.0002137883,0.00001866435,0.002176882,0.0002580432,0.00002774265,0.00005032109,0.00007420221,0.000369046,0.0004800745,0.0008199028,0.9954774,0.00003398465],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00007241987,0.00004926107,0.00002588517,0.00006109192,0.00002856673,0.000004643151,0.9987036,0.0003847951,0.0006698302],"genre_scores_gemma":[0.0002647321,0.00004863127,0.00009405148,0.00004367618,0.00001126559,0.00002376978,0.9985279,0.0001174217,0.0008685131],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.8112509,"threshold_uncertainty_score":0.6314287,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02481491326330058,"score_gpt":0.2606897398731522,"score_spread":0.2358748266098517,"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."}}