{"id":"W4392255248","doi":"10.1145/3638209.3638224","title":"Occupancy Estimation in Smart Buildings: Impact of Data Quality on Feature Selection","year":2023,"lang":"en","type":"article","venue":"","topic":"Mobile Crowdsensing and Crowdsourcing","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Computer science; Feature selection; Reliability (semiconductor); Machine learning; Data mining; Feature (linguistics); Selection (genetic algorithm); Occupancy; Generalization; Artificial intelligence; Quality (philosophy); Process (computing); Power (physics); Engineering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008447189,0.001049616,0.00137829,0.001072518,0.0007740518,0.001225217,0.00111836,0.001043362,0.0005320645],"category_scores_gemma":[0.03589137,0.0003207114,0.0008022945,0.001127077,0.001198193,0.001810861,0.001549631,0.0009945617,0.0001305101],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006519874,"about_ca_system_score_gemma":0.001083817,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008515372,"about_ca_topic_score_gemma":0.004464481,"domain_scores_codex":[0.9954092,0.00168901,0.0003637466,0.0007600466,0.001382516,0.0003954967],"domain_scores_gemma":[0.9778711,0.01687517,0.00127516,0.001510542,0.002103161,0.0003647542],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001658853,0.0005043229,0.08734544,0.0005499262,0.0003516226,0.0004812526,0.0004257794,0.5838246,0.008147179,0.002887189,0.002403758,0.3114201],"study_design_scores_gemma":[0.00006220357,0.0003249484,0.02355419,0.00006967585,0.00007566785,0.0002222333,0.0002163439,0.9639389,0.006613994,0.003844341,0.001037161,0.00004032045],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4809106,0.002390567,0.5125576,0.0009716749,0.0001589982,0.0002234062,0.0004444318,0.0008869474,0.001455681],"genre_scores_gemma":[0.9701303,0.0002072637,0.02876849,0.00009090405,0.00004149379,0.00006918095,0.0004263345,0.00003277471,0.0002332347],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008515372,"threshold_uncertainty_score":0.0446735,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06198769871351883,"score_gpt":0.3803489362575788,"score_spread":0.3183612375440599,"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."}}