{"id":"W3047281291","doi":"10.1007/s40593-020-00210-6","title":"Assessing the Effectiveness of Student Advice Recommender Agent (SARA): the Case of Automated Personalized Feedback","year":2020,"lang":"en","type":"article","venue":"International Journal of Artificial Intelligence in Education","topic":"Online Learning and Analytics","field":"Computer Science","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan","funders":"","keywords":"Mahalanobis distance; Matching (statistics); Propensity score matching; Computer science; Advice (programming); k-nearest neighbors algorithm; Class (philosophy); Recommender system; Machine learning; Artificial intelligence; Statistics; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001789369,0.00008746221,0.0001701253,0.0001338702,0.00005194208,0.000157476,0.0009970996,0.00003271079,0.00001511827],"category_scores_gemma":[0.0005871472,0.00005652549,0.0001105638,0.0004387809,0.00009766606,0.000381815,0.00009553519,0.000269123,0.000003645907],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001025017,"about_ca_system_score_gemma":0.000387908,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001281549,"about_ca_topic_score_gemma":0.00001331881,"domain_scores_codex":[0.9980726,0.0005864398,0.0007312812,0.0001257833,0.0003925884,0.00009131266],"domain_scores_gemma":[0.9971726,0.001008619,0.0007167193,0.0001377978,0.0009220046,0.00004225615],"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":[0.0004723787,0.00365206,0.01511505,0.0001839565,0.0006834272,0.0003228997,0.06500118,0.1370991,0.01461465,0.1343085,0.0008906166,0.6276562],"study_design_scores_gemma":[0.0001784098,0.000286892,0.01578787,0.0006364164,0.0000565779,0.0005978686,0.03506975,0.909355,0.01531583,0.02225886,0.0002741682,0.0001823115],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8888723,0.0001980835,0.08613581,0.0229325,0.001635744,0.000127581,0.000001510448,0.00001814055,0.00007827472],"genre_scores_gemma":[0.9959344,0.00003848563,0.003549116,0.0002518669,0.0002135068,0.000002430276,0.000001253694,0.000004942053,0.000003986439],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.772256,"threshold_uncertainty_score":0.2305043,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0593746033482964,"score_gpt":0.4315131004440099,"score_spread":0.3721384970957134,"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."}}