{"id":"W4362698458","doi":"10.1038/s41598-023-32484-w","title":"Using machine learning to predict student retention from socio-demographic characteristics and app-based engagement metrics","year":2023,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Online Learning and Analytics","field":"Computer Science","cited_by":64,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Predictive power; Dropout (neural networks); Generalizability theory; Mentorship; Computer science; Student engagement; Predictive analytics; Centrality; Predictive validity; Psychology; Macro; Learning analytics; Machine learning; Medical education; Mathematics education; Medicine; Statistics","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":[0.003808688,0.0001726976,0.0002322857,0.0008560916,0.0008112381,0.001015578,0.0003230108,0.0000643173,0.000009417502],"category_scores_gemma":[0.0005174179,0.0001661349,0.00009681176,0.002901575,0.00008306623,0.0001919053,0.0004663883,0.000316979,0.00002086443],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005843127,"about_ca_system_score_gemma":0.000103236,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004891736,"about_ca_topic_score_gemma":0.000005162043,"domain_scores_codex":[0.9969818,0.0001886858,0.0005172055,0.0009686355,0.0009670049,0.0003766812],"domain_scores_gemma":[0.9984493,0.0001193029,0.0003482707,0.0007013472,0.0001768119,0.0002049247],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000005086536,0.000167476,0.9440787,0.00006347216,0.00009099352,0.0008483615,0.002179059,0.0134572,0.01864539,0.0001697038,0.00119542,0.01909919],"study_design_scores_gemma":[0.0001653711,0.00007401058,0.07298263,0.00009044678,0.00005703053,0.00001316561,0.0001731923,0.9141061,0.0004561412,0.001624803,0.009970923,0.0002861327],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9260718,0.00007796304,0.06934544,0.0006551965,0.003124992,0.0002045615,0.000006295477,0.0004909883,0.00002272777],"genre_scores_gemma":[0.9795637,0.00001199291,0.01930585,0.00005384407,0.00009981541,0.000006618262,0.0001484677,0.00001736381,0.0007924017],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.900649,"threshold_uncertainty_score":0.9793238,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0421789404460062,"score_gpt":0.3052658242393649,"score_spread":0.2630868837933587,"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."}}