{"id":"W2293147617","doi":"10.1609/icwsm.v6i5.14223","title":"Using Social Media to Infer Gender Composition of Commuter Populations","year":2021,"lang":"en","type":"article","venue":"Proceedings of the International AAAI Conference on Web and Social Media","topic":"Social Media and Politics","field":"Social Sciences","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Census; Social media; Inference; American Community Survey; Public transport; Microblogging; Ground truth; Geography; Public use; Order (exchange); Service (business); Advertising; Internet privacy; Computer science; Transport engineering; Business; Political science; Sociology; World Wide Web; Marketing; Population; Engineering; Demography; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002132259,0.0001045445,0.0002207862,0.00008475892,0.0004298682,0.00007674262,0.0003259116,0.0001417952,0.0001481605],"category_scores_gemma":[0.0007693755,0.00009729538,0.0000997414,0.0002132855,0.0003820357,0.0001404166,0.0001344783,0.0001696959,0.000002761904],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001020936,"about_ca_system_score_gemma":0.0002847452,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001188274,"about_ca_topic_score_gemma":0.0003891627,"domain_scores_codex":[0.9985879,0.00004360005,0.0002854271,0.000148665,0.0007411799,0.0001932227],"domain_scores_gemma":[0.9983546,0.0002487166,0.0001979904,0.00003383257,0.001085528,0.00007930312],"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":[0.00002519449,0.00008234168,0.01453456,0.00002178685,0.00004139038,3.584421e-7,0.1451751,4.707394e-7,0.02228781,0.8156666,0.0008809048,0.001283499],"study_design_scores_gemma":[0.002418418,0.00006506471,0.2039984,0.0005963884,0.0003847822,0.000007056362,0.3074993,0.0005172037,0.03902989,0.4282882,0.01610244,0.001092802],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9739825,0.00001241177,0.00001068435,0.01021525,0.001424853,0.0001095073,0.00008910846,0.00001588705,0.01413986],"genre_scores_gemma":[0.9977931,0.0000257546,0.0004947612,0.0004890799,0.001121111,0.000007860453,0.00001243205,0.000009018138,0.00004687979],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3873783,"threshold_uncertainty_score":0.3967591,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2027960139324408,"score_gpt":0.3792367506100029,"score_spread":0.1764407366775621,"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."}}