{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.06529381,0.0005364123,0.001460818,0.001889883,0.0009378658,0.00160355,0.00118982,0.001735911,0.001743248],"category_scores_gemma":[0.189691,0.0003254895,0.001429357,0.001496601,0.0009317358,0.002375259,0.001193056,0.001208452,0.0004340535],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001348947,"about_ca_system_score_gemma":0.001778585,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003601681,"about_ca_topic_score_gemma":0.003501408,"domain_scores_codex":[0.9385402,0.04643502,0.003527876,0.003153186,0.007642563,0.0007011875],"domain_scores_gemma":[0.7791088,0.1874433,0.01544469,0.009205612,0.007201816,0.001595711],"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":[0.01232477,0.00763536,0.2740018,0.001610606,0.002575412,0.0002111097,0.004215181,0.01772994,0.002663666,0.002284551,0.001577779,0.6731699],"study_design_scores_gemma":[0.003499059,0.0856078,0.6350046,0.0009724128,0.005756376,0.0008195057,0.006178731,0.2232595,0.01760169,0.009296241,0.01159383,0.0004103241],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.975821,0.001061247,0.01774954,0.0009567017,0.0001019527,0.00103665,0.0001995613,0.0001974691,0.002875938],"genre_scores_gemma":[0.9718038,0.0002237545,0.02663964,0.0001876091,0.00004654514,0.0005244035,0.0001004404,0.00001532358,0.0004584519],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06529381,"threshold_uncertainty_score":0.3453108,"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."}}