{"id":"W4391093649","doi":"10.1109/bigdata59044.2023.10386776","title":"Unstructured Transportation Safety Board Findings Categorization Using the Knowledge Graph Pipeline","year":2023,"lang":"en","type":"article","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Categorization; Computer science; Pipeline (software); Graph; Knowledge graph; Data science; Information retrieval; Artificial intelligence; Theoretical computer science; Programming language","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":[],"consensus_categories":[],"category_scores_codex":[0.001474112,0.0009110915,0.0003254464,0.01212484,0.0007427294,0.00206435,0.00107852,0.001027218,0.002027875],"category_scores_gemma":[0.009276318,0.0002758934,0.001185609,0.006376769,0.0005946276,0.00376467,0.001611463,0.00132307,0.0008996816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001998305,"about_ca_system_score_gemma":0.002194105,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02756493,"about_ca_topic_score_gemma":0.03852771,"domain_scores_codex":[0.9987738,0.0002732536,0.0001215741,0.0003942732,0.0003192027,0.0001179429],"domain_scores_gemma":[0.9947101,0.003258572,0.000485482,0.0006388274,0.0007367631,0.0001702703],"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":[0.0006318259,0.0008681875,0.1163885,0.001382395,0.0004479839,0.002277225,0.002809989,0.1074269,0.01288841,0.02514342,0.03969394,0.6900412],"study_design_scores_gemma":[0.00006802953,0.0002158484,0.04520971,0.0003764499,0.000288651,0.0007229373,0.003660085,0.8245491,0.01011663,0.05781229,0.05688438,0.00009599954],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.373236,0.002365836,0.5303646,0.003377448,0.0003021535,0.001848364,0.05709791,0.01370995,0.0176977],"genre_scores_gemma":[0.6506711,0.0008504734,0.2811911,0.0003121949,0.0000680158,0.0003919512,0.06338374,0.0002232012,0.002908371],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02756493,"threshold_uncertainty_score":0.05480897,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.14575456692883,"score_gpt":0.4031670788603776,"score_spread":0.2574125119315476,"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."}}