{"id":"W7132985241","doi":"","title":"Language Diversity in Toronto Assignment, Step 2: Data Elicitation","year":2019,"lang":"en","type":"other","venue":"TSpace","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Diversity (politics); Natural language; Linguistic diversity; Language identification; Data collection; Computational linguistics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00649151,0.0006741809,0.0007686943,0.00386711,0.003299901,0.002305466,0.001797392,0.001765147,0.08031038],"category_scores_gemma":[0.03772891,0.0007854881,0.0006974346,0.002647978,0.0005576382,0.001269539,0.003683302,0.0016473,0.02573089],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005865533,"about_ca_system_score_gemma":0.01972786,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.08649257,"about_ca_topic_score_gemma":0.1735581,"domain_scores_codex":[0.9957516,0.001086352,0.000721813,0.0006195419,0.0009974118,0.0008232003],"domain_scores_gemma":[0.9798535,0.006359049,0.001572861,0.001427806,0.008641167,0.002145632],"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":[0.002031567,0.001714998,0.08165973,0.004481469,0.00005186066,0.00153764,0.03330161,0.001655549,0.004579783,0.007701674,0.731032,0.1302521],"study_design_scores_gemma":[0.0007010391,0.0005825208,0.1799301,0.003093426,0.0001174784,0.0003992131,0.04469977,0.004172555,0.008565666,0.005959466,0.7515543,0.0002244884],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"other","genre_scores_codex":[0.1621301,0.0004409543,0.02988974,0.005653406,0.0006035241,0.1806758,0.5161219,0.001471456,0.103013],"genre_scores_gemma":[0.1829176,0.000578952,0.06409317,0.002837056,0.0002702399,0.4648077,0.1893528,0.000555712,0.09458661],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.9135075,"threshold_uncertainty_score":0.268665,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05506345764781739,"score_gpt":0.3734721393750575,"score_spread":0.3184086817272401,"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."}}