{"id":"W4411120873","doi":"10.1609/icwsm.v19i1.35940","title":"UKTwitNewsCor: A Dataset of Online Local News Articles for the Study of Local News Provision","year":2025,"lang":"en","type":"article","venue":"Proceedings of the International AAAI Conference on Web and Social Media","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Surrey Place Centre","funders":"","keywords":"Political science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"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":[0.0007955484,0.0008274131,0.0007234825,0.008244905,0.001126288,0.001951325,0.001315368,0.001675177,0.01177785],"category_scores_gemma":[0.00881627,0.0003630843,0.0005910002,0.01138738,0.0005584235,0.002185051,0.002293858,0.0008259108,0.01756155],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001285936,"about_ca_system_score_gemma":0.001444444,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02761536,"about_ca_topic_score_gemma":0.07646593,"domain_scores_codex":[0.9986866,0.0002996938,0.0002085232,0.0003035254,0.0003424243,0.0001592248],"domain_scores_gemma":[0.9944006,0.001928797,0.0008513831,0.00090273,0.001417464,0.0004990051],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003731547,0.0002124273,0.02903714,0.002797885,0.0001158855,0.001100762,0.002330448,0.001100176,0.002746033,0.002875194,0.9091486,0.04816226],"study_design_scores_gemma":[0.0001049578,0.00008634188,0.06095019,0.0004062142,0.00007095675,0.0005929334,0.002769015,0.003605511,0.002130585,0.001277158,0.9279234,0.00008268498],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.03082542,0.0008900769,0.001867708,0.0004893932,0.0001887004,0.0002380862,0.9552544,0.001241406,0.009004766],"genre_scores_gemma":[0.01799567,0.0002958429,0.004469526,0.0001162967,0.00006536883,0.0005362799,0.9725729,0.0001664802,0.003781621],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.02761536,"threshold_uncertainty_score":0.05490923,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07310951934286138,"score_gpt":0.3834048035734087,"score_spread":0.3102952842305473,"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."}}