{"id":"W4406934208","doi":"10.7717/peerj.18853","title":"Evaluating the feasibility of automating dataset retrieval for biodiversity monitoring","year":2025,"lang":"en","type":"article","venue":"PeerJ","topic":"Research Data Management Practices","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke","funders":"","keywords":"Computer science; Relevance (law); Ranking (information retrieval); Information retrieval; Overfitting; Biodiversity; Database; Pace; Data mining; Data science; Machine learning; Geography; Ecology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.0175387,0.001264828,0.0008006464,0.007525599,0.001012551,0.003225152,0.001827045,0.001920665,0.001513874],"category_scores_gemma":[0.05279205,0.0003428449,0.001205265,0.004032108,0.0005610962,0.003546364,0.002133891,0.0008064954,0.001339897],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001260982,"about_ca_system_score_gemma":0.002575342,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.010602,"about_ca_topic_score_gemma":0.009412301,"domain_scores_codex":[0.9876668,0.004553966,0.0024328,0.002017098,0.002838781,0.0004904366],"domain_scores_gemma":[0.9387272,0.04358096,0.003471711,0.004149664,0.009136287,0.0009342013],"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":[0.006007341,0.004198201,0.1574201,0.006437356,0.001179147,0.0008171413,0.002277836,0.02141775,0.0689114,0.001093832,0.02360058,0.7066393],"study_design_scores_gemma":[0.001307869,0.005135189,0.2845796,0.0007956316,0.001300062,0.001486548,0.002618871,0.5277885,0.1386424,0.002108921,0.03383307,0.0004033784],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9093184,0.001844798,0.05081008,0.0009101143,0.0002194989,0.002139347,0.006635603,0.02409583,0.004026235],"genre_scores_gemma":[0.7991842,0.0006408195,0.1802506,0.0002859042,0.00009210956,0.0009263437,0.01640774,0.0005376837,0.001674671],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9824613,"threshold_uncertainty_score":0.0927546,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4107494996393343,"score_gpt":0.5163996542726087,"score_spread":0.1056501546332744,"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."}}