{"id":"W3041699688","doi":"10.1371/journal.pone.0239441","title":"Public discourse and sentiment during the COVID 19 pandemic: Using Latent Dirichlet Allocation for topic modeling on Twitter","year":2020,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Misinformation and Its Impacts","field":"Social Sciences","cited_by":355,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"University of Toronto; National Natural Science Foundation of China","keywords":"Latent Dirichlet allocation; Topic model; Sentiment analysis; Pandemic; Coronavirus disease 2019 (COVID-19); Social media; Outbreak; Public health; Coronavirus; Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); 2019-20 coronavirus outbreak; Cruise; Geography; Psychology; Medicine; History; Computer science; Natural language processing; Virology; Pathology; World Wide Web","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":[],"consensus_categories":[],"category_scores_codex":[0.0002705378,0.00005244248,0.00007113285,0.00002336164,0.0004351964,0.0001367777,0.00007295693,0.0000288797,0.00004770778],"category_scores_gemma":[0.0002774953,0.00003827156,0.00001892806,0.00007716334,0.00003488134,0.0002286049,0.00002343696,0.0000554555,0.000005853814],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008727697,"about_ca_system_score_gemma":0.00006170387,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006670957,"about_ca_topic_score_gemma":0.00002845164,"domain_scores_codex":[0.999335,0.00004199696,0.0001208498,0.00009056913,0.0002549482,0.0001566282],"domain_scores_gemma":[0.9996429,0.0000376797,0.00004904468,0.00006410407,0.0000400849,0.000166196],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003071275,0.001266357,0.07637691,0.00101125,0.0006712683,0.000002607517,0.8381107,0.02211167,0.0315887,0.02095338,0.001749601,0.005850392],"study_design_scores_gemma":[0.002123825,0.0001378471,0.002256759,0.000131959,0.0001831549,0.000001604481,0.02669701,0.9621638,0.001298016,0.0009941006,0.00352756,0.0004843401],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.954944,0.00002668856,0.003265445,0.04087272,0.00001997184,0.0003676708,0.000003208262,0.00003774297,0.0004625564],"genre_scores_gemma":[0.9948651,0.0000634675,0.0002511877,0.00441801,0.0001607135,0.000005703618,0.000004979986,0.000004440585,0.0002263626],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9400522,"threshold_uncertainty_score":0.3347223,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4339952009345251,"score_gpt":0.3687804757839991,"score_spread":0.06521472515052595,"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."}}