{"id":"W4398139524","doi":"10.1177/15210251241254053","title":"Educational Data Mining in Higher Education: Building a Predictive Model for Retaining University Graduates as Master's Students","year":2024,"lang":"en","type":"article","venue":"Journal of College Student Retention Research Theory & Practice","topic":"Online Learning and Analytics","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Brock University","funders":"","keywords":"Residence; Medical education; Value (mathematics); Higher education; Predictive value; Quality (philosophy); Institution; Psychology; Mathematics education; Computer science; Sociology; Medicine; Political science; Machine learning; Social 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01390278,0.0001313136,0.0002114995,0.0009763815,0.0003426797,0.0006428412,0.002249727,0.00006696913,0.00002677686],"category_scores_gemma":[0.002302119,0.0001243012,0.0001036455,0.001230035,0.00008134179,0.003409431,0.0009011469,0.0008572878,0.000007374847],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004368011,"about_ca_system_score_gemma":0.002594619,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007197818,"about_ca_topic_score_gemma":0.000002404341,"domain_scores_codex":[0.9957104,0.001358469,0.0004909527,0.0004501483,0.001667195,0.0003228218],"domain_scores_gemma":[0.9939989,0.003655363,0.0003397851,0.0004613916,0.001389685,0.0001548323],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000934885,0.001410374,0.003247132,0.0001307167,0.0006238697,0.0002180202,0.003977855,0.003316349,0.000141714,0.971121,0.006636743,0.008241417],"study_design_scores_gemma":[0.002128009,0.001454208,0.008980569,0.002495841,0.000386073,0.0003871247,0.03566865,0.6790703,0.0000189865,0.2125187,0.05642752,0.0004640263],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4807771,0.005418628,0.2744675,0.2207711,0.00412542,0.001838147,0.00009809359,0.0001316582,0.0123723],"genre_scores_gemma":[0.9381014,0.0002577647,0.03937607,0.0001079183,0.0004603094,0.000006388376,0.000009266993,0.00001605115,0.0216649],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7586023,"threshold_uncertainty_score":0.6198933,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1632851712291555,"score_gpt":0.475428385306471,"score_spread":0.3121432140773156,"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."}}