{"id":"W4229069564","doi":"10.2196/35266","title":"Enhancing COVID-19 Epidemic Forecasting Accuracy by Combining Real-time and Historical Data From Multiple Internet-Based Sources: Analysis of Social Media Data, Online News Articles, and Search Queries","year":2022,"lang":"en","type":"article","venue":"JMIR Public Health and Surveillance","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"City University of Hong Kong","keywords":"The Internet; Computer science; Social media; Data science; Big data; Data mining; World Wide Web","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.003867663,0.0002553379,0.001241596,0.0003333075,0.0004211387,0.00007140251,0.0005192356,0.00007671645,0.0001644952],"category_scores_gemma":[0.009903823,0.0002575397,0.00005131787,0.0009428512,0.0002534458,0.000321833,0.001378222,0.0004228833,7.587363e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004446429,"about_ca_system_score_gemma":0.001463316,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01619808,"about_ca_topic_score_gemma":0.01503678,"domain_scores_codex":[0.9954022,0.001239171,0.001018857,0.001077566,0.0006187725,0.0006434905],"domain_scores_gemma":[0.9912836,0.005726172,0.0005539565,0.001069837,0.0001026685,0.001263783],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006296718,0.0002943058,0.9468571,0.0004307908,0.0003076968,0.00002031491,0.003742435,0.00001249313,0.0003901028,0.000007160083,0.02603799,0.02126996],"study_design_scores_gemma":[0.005468315,0.000349092,0.28433,0.00004427098,0.00007913286,0.00001948786,0.005791187,0.544036,0.000004006456,0.00001129508,0.1593185,0.0005487107],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9664738,0.003360276,0.001198134,0.008661604,0.00006329943,0.0004644588,0.01964993,0.000119007,0.000009558348],"genre_scores_gemma":[0.9402763,0.0003721432,0.001556683,0.00222179,0.0001045219,0.00003007316,0.05536936,0.00003345213,0.00003563388],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6625271,"threshold_uncertainty_score":0.9999877,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1603240612307615,"score_gpt":0.369769992526947,"score_spread":0.2094459312961855,"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."}}