{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001779959,0.002483818,0.001303448,0.01012989,0.001233267,0.003557588,0.001815809,0.00170873,0.01210419],"category_scores_gemma":[0.007832821,0.001318903,0.004142971,0.005093119,0.0005646585,0.005137239,0.003547941,0.001996156,0.007533046],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002093275,"about_ca_system_score_gemma":0.002800561,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01804214,"about_ca_topic_score_gemma":0.03498783,"domain_scores_codex":[0.9983068,0.0002472202,0.0004136566,0.0005791762,0.0003497448,0.0001034012],"domain_scores_gemma":[0.9977437,0.00107273,0.0002673504,0.0004656071,0.0003602216,0.00009028926],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001234416,0.0006868558,0.03224066,0.008314868,0.0009489989,0.003457312,0.005196729,0.0173612,0.03837465,0.03105233,0.234657,0.626475],"study_design_scores_gemma":[0.0002912727,0.0001975187,0.01670642,0.0009622631,0.0004115541,0.001918997,0.001870322,0.3163435,0.03135802,0.02888452,0.6007677,0.0002878392],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04061951,0.003405376,0.4169669,0.001472415,0.0005156109,0.001641901,0.2634618,0.2606799,0.01123654],"genre_scores_gemma":[0.0549757,0.0006668357,0.6786078,0.0004379916,0.00004172647,0.00144038,0.2572157,0.003291355,0.003322453],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01804214,"threshold_uncertainty_score":0.04049253,"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."}}