{"id":"W4415320279","doi":"10.48550/arxiv.2505.14804","title":"Automated Journalistic Questions: A New Method for Extracting 5W1H in French","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Pipeline (software); Task (project management); Journalism; Information extraction; Task analysis","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.001966056,0.001677924,0.0007078507,0.008361829,0.002130102,0.002964222,0.000956262,0.001935921,0.01186756],"category_scores_gemma":[0.005118319,0.0007504938,0.001563127,0.002932256,0.001157312,0.003545196,0.002027326,0.001535772,0.005062083],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002049766,"about_ca_system_score_gemma":0.003833398,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03444337,"about_ca_topic_score_gemma":0.04245107,"domain_scores_codex":[0.9980019,0.0005102393,0.0001853372,0.0006988833,0.0004280275,0.0001756077],"domain_scores_gemma":[0.9972081,0.001270771,0.0002400681,0.000269108,0.0009079226,0.0001040805],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000229633,0.000152994,0.005287461,0.001184048,0.000135913,0.0009599192,0.004632199,0.00332158,0.05394482,0.01999719,0.06088927,0.849265],"study_design_scores_gemma":[0.0001696193,0.0003115825,0.02332429,0.0003440537,0.0002916205,0.003359372,0.004267677,0.2228366,0.09006756,0.04262231,0.6120709,0.0003344986],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03903878,0.001493012,0.8790445,0.002509667,0.000334477,0.00109495,0.01766764,0.04274726,0.01606974],"genre_scores_gemma":[0.1273676,0.0005873096,0.8292653,0.0004859692,0.0003542859,0.0006466745,0.0282372,0.001822735,0.01123299],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.03444337,"threshold_uncertainty_score":0.0684858,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05413978131493793,"score_gpt":0.401088743874645,"score_spread":0.3469489625597071,"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."}}