{"id":"W4404619910","doi":"10.3897/biss.8.142382","title":"Relation Extraction From Unstructured Species Descriptions Using TaxonNERD and LLaMA 2 7B","year":2024,"lang":"en","type":"article","venue":"Biodiversity Information Science and Standards","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Ministerio de Ciencia Tecnología y Telecomunicaciones; Instituto Tecnológico de Costa Rica; Consejo Superior Universitario Centroamericano; International Development Research Centre","keywords":"Relationship extraction; Adaptability; Computer science; Relation (database); Biodiversity; Taxonomy (biology); False positive paradox; Trophic level; Artificial intelligence; Natural language processing; Information extraction; Information retrieval; Ecology; Biology; Data mining","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004475851,0.00006181301,0.00005272285,0.000115898,0.0003901866,0.0002604812,0.00006092937,0.00008564718,0.00001854327],"category_scores_gemma":[0.0002684563,0.00005271144,0.00001579032,0.0001893756,0.0004960626,0.0001341821,0.00007404605,0.00005493486,0.00000380328],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006863488,"about_ca_system_score_gemma":0.0002075932,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004230023,"about_ca_topic_score_gemma":0.00001072103,"domain_scores_codex":[0.9992709,0.00001039406,0.00009843957,0.0001324597,0.0003885387,0.00009932237],"domain_scores_gemma":[0.9995619,0.000008822063,0.00003423798,0.00006461475,0.0002713489,0.00005906526],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003357245,0.00002727405,0.03045275,0.0001576577,0.0001258435,0.00001000177,0.01158053,0.0001761993,0.4241205,0.001593798,0.08014382,0.4512759],"study_design_scores_gemma":[0.0004186724,0.0001490519,0.05991798,0.00003880523,0.00004110492,0.00003792271,0.005409857,0.005104115,0.01234007,0.0001366143,0.9161595,0.0002463034],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9859608,0.0003442556,0.01184354,0.0002957127,0.0003733775,0.00005483392,0.000543958,0.00002792139,0.0005555442],"genre_scores_gemma":[0.9975539,0.0002972657,0.001923945,0.00009502353,0.00003812736,5.100267e-7,0.00007520974,6.157156e-7,0.00001537924],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8360157,"threshold_uncertainty_score":0.300104,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02713421106529408,"score_gpt":0.279478457602686,"score_spread":0.2523442465373919,"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."}}