{"id":"W4412366363","doi":"10.1080/10875301.2025.2529435","title":"Evaluating the Impact of a Proactive Chat Feature on Reference Questions, Their Complexity and Subject Matter","year":2025,"lang":"en","type":"article","venue":"Internet Reference Services Quarterly","topic":"Digital Communication and Language","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Feature (linguistics); Computer science; Subject matter; Subject (documents); World Wide Web; Psychology; Information retrieval; Data science; Linguistics","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":[0.03184881,0.000550208,0.0007918238,0.002559291,0.001252104,0.00467043,0.001193105,0.0009654859,0.003586869],"category_scores_gemma":[0.2134588,0.0004749969,0.0007186658,0.001498601,0.001083351,0.0038137,0.003083818,0.001316785,0.0008273902],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00215927,"about_ca_system_score_gemma":0.002744932,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002076287,"about_ca_topic_score_gemma":0.005232831,"domain_scores_codex":[0.9665976,0.02179045,0.002156301,0.001279754,0.006956207,0.001219764],"domain_scores_gemma":[0.532068,0.4049663,0.02373741,0.009562738,0.02289464,0.006770955],"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":[0.003704217,0.004422674,0.3430753,0.003446638,0.0002497533,0.001163218,0.150858,0.00126923,0.03158621,0.001118526,0.002257033,0.4568491],"study_design_scores_gemma":[0.0002226037,0.01418369,0.868197,0.001032176,0.0005464712,0.0009170966,0.07912001,0.003426208,0.01449131,0.0009270119,0.0166479,0.0002885328],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9925445,0.0002140319,0.002442376,0.0003427288,0.00002701232,0.0003936622,0.00006804199,0.0002572201,0.003710438],"genre_scores_gemma":[0.987327,0.0001795747,0.00967146,0.0002016024,0.00004492724,0.0004306831,0.0001372075,0.00006651726,0.001940939],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03184881,"threshold_uncertainty_score":0.1684346,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06027548121335283,"score_gpt":0.3701431148807869,"score_spread":0.3098676336674341,"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."}}