{"id":"W4381333253","doi":"10.12688/openreseurope.16003.1","title":"Investigating antiquities trafficking with generative pre-trained transformer (GPT)-3 enabled knowledge graphs: A case study","year":2023,"lang":"en","type":"article","venue":"Open Research Europe","topic":"Archaeological Research and Protection","field":"Earth and Planetary Sciences","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"Social Sciences and Humanities Research Council of Canada; Horizon 2020 Framework Programme; European Commission","keywords":"Computer science; Embedding; Newspaper; Python (programming language); Artificial intelligence; Snapshot (computer storage); Natural language processing; Theoretical computer science; Programming language; Database","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.006496632,0.0002186854,0.0002832863,0.0004486562,0.001951263,0.0006215278,0.000824099,0.00006073316,0.001146731],"category_scores_gemma":[0.0009524764,0.0001410655,0.00004118166,0.003548343,0.0007390928,0.000704582,0.0001732143,0.0009242834,0.0004267507],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001087406,"about_ca_system_score_gemma":0.0004471951,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.01175219,"about_ca_topic_score_gemma":0.04179618,"domain_scores_codex":[0.9941677,0.002770905,0.0002992645,0.000673193,0.0009348212,0.001154142],"domain_scores_gemma":[0.9976756,0.001069849,0.0000462798,0.0003002856,0.0004233251,0.0004846678],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001919703,0.0004106657,0.1597825,0.0003333951,0.0004340853,0.02715987,0.1260938,0.0047941,0.003187787,0.001534952,0.003117815,0.6712313],"study_design_scores_gemma":[0.01137275,0.05317398,0.5836025,0.0007589929,0.00008441474,0.00299561,0.1945229,0.09620137,0.005703292,0.01560633,0.03316937,0.002808456],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9683076,0.0001700251,0.0001051161,0.0004608881,0.00004017275,0.003053544,0.00003679592,0.0001326433,0.02769327],"genre_scores_gemma":[0.9944966,0.0001590293,0.0007745459,0.00003676672,0.00008491767,0.0001011457,0.00005508076,0.00001612006,0.004275765],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6684229,"threshold_uncertainty_score":0.9997663,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2018112595288598,"score_gpt":0.4021518027682092,"score_spread":0.2003405432393494,"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."}}