{"id":"W4392473475","doi":"10.2196/49411","title":"Machine Learning–Based Approach for Identifying Research Gaps: COVID-19 as a Case Study","year":2024,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Academic Publishing and Open Access","field":"Decision Sciences","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Cluster analysis; Relevance (law); Data science; Artificial intelligence; Coronavirus disease 2019 (COVID-19); Scalability; Information retrieval; Machine learning; Scientific literature; Data mining","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch","bibliometrics"],"consensus_categories":[],"category_scores_codex":[0.05747543,0.0008022182,0.001493493,0.0291959,0.003559196,0.005641678,0.002671405,0.003565623,0.004349979],"category_scores_gemma":[0.100598,0.0005469801,0.002534806,0.03436713,0.001721783,0.004175455,0.004032603,0.001543853,0.0006926325],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.007439265,"about_ca_system_score_gemma":0.01448614,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007733968,"about_ca_topic_score_gemma":0.01090895,"domain_scores_codex":[0.958275,0.02516935,0.006951316,0.002346355,0.00611082,0.001147168],"domain_scores_gemma":[0.7976575,0.1695124,0.0130317,0.003886749,0.01339292,0.002518702],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00196388,0.00172401,0.1764481,0.07501915,0.001855891,0.03862099,0.05738506,0.01593084,0.005681801,0.0733958,0.04928901,0.5026856],"study_design_scores_gemma":[0.001183418,0.001974747,0.1009002,0.05453905,0.004704923,0.01372856,0.1320288,0.06063691,0.01306626,0.07839584,0.53825,0.000591352],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5477266,0.1085137,0.1814131,0.05120049,0.001138975,0.02192309,0.03015555,0.0008843639,0.0570441],"genre_scores_gemma":[0.638037,0.01759879,0.3175471,0.003130698,0.0003234569,0.01103064,0.008729601,0.00009700375,0.003505679],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9708041,"threshold_uncertainty_score":0.3039628,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5411560199070763,"score_gpt":0.6402366472363406,"score_spread":0.09908062732926426,"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."}}