{"id":"W4409456724","doi":"10.3390/bdcc9040099","title":"Predicting College Enrollment for Low-Socioeconomic-Status Students Using Machine Learning Approaches","year":2025,"lang":"en","type":"article","venue":"Big Data and Cognitive Computing","topic":"Online Learning and Analytics","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; University of Alberta","funders":"","keywords":"Socioeconomic status; Mathematics education; Psychology; Computer science; Medical education; Machine learning; Artificial intelligence; Demography; Medicine; Sociology","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":[0.0007328212,0.0001984554,0.00028311,0.0001508221,0.0007235036,0.0003478482,0.0007401698,0.00006183405,6.678924e-7],"category_scores_gemma":[0.0003505569,0.0002041389,0.00004677755,0.0002057883,0.00005845268,0.0002377642,0.00220285,0.0002706456,0.000002060215],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004388539,"about_ca_system_score_gemma":0.0001339921,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004285089,"about_ca_topic_score_gemma":0.00001320783,"domain_scores_codex":[0.9982247,0.0001231716,0.0003377914,0.0007335303,0.0001552097,0.0004255358],"domain_scores_gemma":[0.9986619,0.0006624363,0.0001970503,0.0002961401,0.0000938477,0.00008860262],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003035953,0.0001681188,0.4553443,0.000179622,0.0003027396,0.000007759855,0.0008063564,0.00125258,0.00004887997,0.001499389,0.00003082895,0.5403291],"study_design_scores_gemma":[0.001235822,0.00005940617,0.005925132,0.0003127145,0.00008379474,0.000004580711,0.001072927,0.9901694,0.00006415934,0.0002610527,0.0006096646,0.0002013677],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4266863,0.0004537775,0.571769,0.0001976663,0.0002526763,0.0002350672,0.0001457925,0.00008947836,0.0001703058],"genre_scores_gemma":[0.9841567,0.00002577798,0.01501541,0.0002212361,0.0002159766,0.00000363383,0.0002017063,0.00001304118,0.0001464725],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9889168,"threshold_uncertainty_score":0.832454,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08870611740529777,"score_gpt":0.3334209559600772,"score_spread":0.2447148385547794,"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."}}