{"id":"W2252149878","doi":"10.3115/v1/w14-2503","title":"Sociolinguistics for Computational Social Science","year":2014,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Sociolinguistics; Computer science; Linguistics","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":[],"consensus_categories":[],"category_scores_codex":[0.004050906,0.001397277,0.001889997,0.006054236,0.002002691,0.006480949,0.00281014,0.002110967,0.02296973],"category_scores_gemma":[0.01235508,0.0005862492,0.002237475,0.005594732,0.007393308,0.007845652,0.004580523,0.004157028,0.006879118],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002841662,"about_ca_system_score_gemma":0.002372054,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004158155,"about_ca_topic_score_gemma":0.002981391,"domain_scores_codex":[0.997329,0.001395692,0.0001980397,0.0005513777,0.0004468121,0.00007902096],"domain_scores_gemma":[0.991811,0.005733011,0.0004784334,0.001256579,0.0004684286,0.0002524546],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00001154391,0.00003137128,0.0005565159,0.0003969737,0.00005593767,0.0001577246,0.0006815384,0.00234728,0.0001601458,0.9436708,0.009593589,0.04233648],"study_design_scores_gemma":[0.000004732292,0.000006658268,0.0002312006,0.00009532023,0.000006545515,0.00005305036,0.0001740457,0.005413183,0.00003064521,0.9465444,0.04743013,0.0000101658],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.006830429,0.0489458,0.8318307,0.02459771,0.002788314,0.000429835,0.003293991,0.002481565,0.07880167],"genre_scores_gemma":[0.2106048,0.0290509,0.7186162,0.004518841,0.006087763,0.001864088,0.005772974,0.0008537693,0.02263065],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02296973,"threshold_uncertainty_score":0.07684135,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01538295941453947,"score_gpt":0.3192795696701142,"score_spread":0.3038966102555747,"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."}}