{"id":"W3034470301","doi":"10.1016/j.forsciint.2020.110364","title":"Social media forensics applied to assessment of post–critical incident social reaction: The case of the 2017 Manchester Arena terrorist attack","year":2020,"lang":"en","type":"article","venue":"Forensic Science International","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":23,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Montréal; Université du Québec à Trois-Rivières","funders":"Economic and Social Research Council","keywords":"Latent Dirichlet allocation; Social media; Law enforcement; Terrorism; Computer science; Data science; Topic model; Set (abstract data type); Event (particle physics); Computational sociology; Perception; Digital forensics; Computer security; Internet privacy; World Wide Web; Artificial intelligence; Political science; Law; Psychology","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001263302,0.00009813211,0.0001799094,0.00008099398,0.0008948811,0.0000924079,0.0008191608,0.0000466547,0.0001008099],"category_scores_gemma":[0.001490141,0.000065423,0.000168062,0.0007993705,0.002234282,0.0001903494,0.000316381,0.0001604848,0.000006355721],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000172012,"about_ca_system_score_gemma":0.0004755954,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002480266,"about_ca_topic_score_gemma":0.001673727,"domain_scores_codex":[0.9976832,0.0001225944,0.0003668635,0.0002710422,0.001345995,0.0002103071],"domain_scores_gemma":[0.998012,0.0006789398,0.0002387685,0.0001145903,0.0008559932,0.00009972517],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.00007327194,0.0001108957,0.005333135,0.00001641284,0.0000965334,0.00002663713,0.09981826,0.0002780766,0.00294997,0.8316269,0.006631658,0.05303826],"study_design_scores_gemma":[0.0009838896,0.0001616218,0.8241169,0.00005888437,0.0002802479,0.00008250863,0.07057107,0.01126947,0.002980193,0.07641232,0.01238681,0.0006960595],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8947524,0.000004063885,0.003420217,0.08772933,0.001430443,0.0002309664,0.00004736887,0.00001532153,0.01236986],"genre_scores_gemma":[0.9954017,4.859766e-7,0.002741016,0.0008815258,0.0009230687,0.00001153008,0.000005418493,0.00000456852,0.0000306207],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8187838,"threshold_uncertainty_score":0.8232305,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1042927051340976,"score_gpt":0.4482298001236109,"score_spread":0.3439370949895133,"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."}}