{"id":"W4402423794","doi":"10.24908/iqurcp18001","title":"Data Mining Virtual Contaminated Sites to Develop an Educational Learning Tool supported by a k-NN Machine Learning Predictive Algorithm for use in Environmental Engineering Education","year":2024,"lang":"en","type":"article","venue":"Inquiry Queen s Undergraduate Research Conference Proceedings","topic":"Data Stream Mining Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Computer science; Machine learning; Artificial intelligence; Algorithm","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.002860324,0.0003812887,0.0003209354,0.00119195,0.0003091297,0.002191912,0.002048148,0.0001711667,0.00001341961],"category_scores_gemma":[0.002701556,0.0004226039,0.00002995361,0.001752798,0.0001718818,0.005542104,0.001618767,0.001076763,0.00002503235],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006576785,"about_ca_system_score_gemma":0.001594653,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002545376,"about_ca_topic_score_gemma":0.00002353366,"domain_scores_codex":[0.9957358,0.0001499116,0.0005904656,0.001642096,0.0009546127,0.0009270766],"domain_scores_gemma":[0.9974802,0.0009778668,0.0001099425,0.0004541234,0.0006534092,0.0003245154],"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":[0.0001353944,0.0009387318,0.02113247,0.0003620385,0.0002352971,0.00003384767,0.02478868,0.00008338451,0.07575364,0.03450297,0.03166854,0.810365],"study_design_scores_gemma":[0.0002875428,0.001003916,0.001071095,0.0006449129,0.00001324638,0.00003883491,0.002352783,0.980944,0.003184045,0.001442025,0.008445642,0.0005719254],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2794349,0.0001194773,0.7134809,0.003639435,0.0002450754,0.001554647,0.0003774595,0.00109967,0.00004844538],"genre_scores_gemma":[0.7982416,0.0001063377,0.1975966,0.00004013407,0.0001306315,0.0006154653,0.002586996,0.00007610525,0.0006061519],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9808607,"threshold_uncertainty_score":0.9998226,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0783110474724833,"score_gpt":0.3578740208104614,"score_spread":0.2795629733379781,"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."}}