{"id":"W4288060620","doi":"10.18357/kula.221","title":"Re-purposing Excavation Database Content as Paradata","year":2022,"lang":"en","type":"article","venue":"KULA knowledge creation dissemination and preservation studies","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Riksbankens Jubileumsfond; European Commission","keywords":"Metadata; Computer science; Documentation; Field (mathematics); Identification (biology); Structuring; Process (computing); Data science; Excavation; Reading (process); Plan (archaeology); Archaeology; World Wide Web","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008853648,0.0001685478,0.0001861987,0.0002743565,0.0009314899,0.0002320882,0.0004451334,0.00003669574,0.00004744448],"category_scores_gemma":[0.001127928,0.0001644482,0.00003181324,0.0007182155,0.00006288107,0.001655055,0.000875639,0.0001649791,0.000006672204],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001525977,"about_ca_system_score_gemma":0.0000471809,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002557004,"about_ca_topic_score_gemma":0.00002566129,"domain_scores_codex":[0.998383,0.0002277743,0.0003623537,0.0004554774,0.0004079248,0.0001635034],"domain_scores_gemma":[0.998253,0.0004442662,0.000253287,0.0004124829,0.0005835397,0.00005337891],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00009055255,0.0004162712,0.005746871,0.0003610594,0.0001576399,0.00001077672,0.07850637,0.00003008648,0.02097802,0.6397868,0.05722685,0.1966887],"study_design_scores_gemma":[0.003281215,0.0008243387,0.01983027,0.001206126,0.0002737542,0.0001168111,0.04420131,0.1550691,0.1607727,0.1834892,0.427934,0.003001125],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1192133,0.1293786,0.6815515,0.03244331,0.001677303,0.003128272,0.0001175897,0.003198477,0.02929171],"genre_scores_gemma":[0.9314791,0.0007081588,0.04443895,0.0004362857,0.00008773011,0.0005881104,0.0005760767,0.00002107862,0.02166451],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8122658,"threshold_uncertainty_score":0.7164361,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09820098497722425,"score_gpt":0.3918604070508239,"score_spread":0.2936594220735996,"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."}}