{"id":"W3120906435","doi":"10.1140/epjds/s13688-020-00254-7","title":"Challenges when identifying migration from geo-located Twitter data","year":2021,"lang":"en","type":"article","venue":"EPJ Data Science","topic":"Human Mobility and Location-Based Analysis","field":"Social Sciences","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"McGill University","keywords":"Geolocation; Suspect; Data science; Computer science; Set (abstract data type); Social media; Reliability (semiconductor); TRACE (psycholinguistics); Data set; Internet privacy; World Wide Web; Sociology; Artificial intelligence","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003804253,0.00009222946,0.0001358465,0.00009680862,0.001213667,0.0006758654,0.003739542,0.00005677488,0.00125949],"category_scores_gemma":[0.002275284,0.00009347274,0.00002114806,0.0008518497,0.0008491374,0.003418087,0.001157908,0.0001299606,0.0002777143],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008213813,"about_ca_system_score_gemma":0.001075391,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.03129006,"about_ca_topic_score_gemma":0.2354736,"domain_scores_codex":[0.9970286,0.0002679921,0.0002431919,0.001170118,0.000958243,0.0003318301],"domain_scores_gemma":[0.9961748,0.0002261616,0.00009603487,0.003080101,0.0002739007,0.0001490727],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00002592591,0.001077824,0.0292879,0.00011583,0.0002528302,0.0001107396,0.2786119,0.0001900965,0.03961062,0.01672472,0.0637834,0.5702082],"study_design_scores_gemma":[0.0005689099,0.00002514098,0.09503257,0.0002359359,0.0003043553,0.000002382132,0.1223002,0.09239521,0.004112286,0.01800826,0.6658481,0.001166614],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7243405,0.01405843,0.06952718,0.1506301,0.003182348,0.0009121099,0.002471476,0.0006779752,0.03419985],"genre_scores_gemma":[0.9929035,0.0009644776,0.002121819,0.0004741693,0.0003479357,0.00000432662,0.002562579,0.000005427389,0.0006157674],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6020647,"threshold_uncertainty_score":0.9996535,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3333258322916927,"score_gpt":0.4153275156938945,"score_spread":0.08200168340220182,"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."}}