{"id":"W4411987470","doi":"10.1016/j.knosys.2025.114008","title":"A heterogeneous federated learning framework for human activity recognition","year":2025,"lang":"en","type":"article","venue":"Knowledge-Based Systems","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; China Scholarship Council","keywords":"Computer science; Human–computer interaction; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":true,"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":["metaresearch","metaepi_narrow","open_science"],"consensus_categories":[],"category_scores_codex":[0.000822065,0.0002669522,0.0003718958,0.0003620007,0.0007284164,0.000612654,0.007374117,0.0003759196,0.000004731253],"category_scores_gemma":[0.01205579,0.000276745,0.0001272844,0.0009099605,0.00007209591,0.0002750149,0.006802453,0.0004918884,0.00007487935],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002915849,"about_ca_system_score_gemma":0.0002038649,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007072854,"about_ca_topic_score_gemma":0.00005013197,"domain_scores_codex":[0.9977925,0.0003705739,0.0003671916,0.0008002153,0.000179365,0.000490178],"domain_scores_gemma":[0.9950211,0.0009676991,0.0002163837,0.003396548,0.0003313967,0.00006688873],"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":[0.0003102368,0.002160437,0.003494825,0.005355146,0.0008132327,0.00008712657,0.0004332855,0.00218408,0.05295074,0.03828174,0.1661879,0.7277413],"study_design_scores_gemma":[0.000944457,0.0002718019,0.00009134775,0.001297923,0.00002735476,0.000007203527,0.00003130998,0.8035602,0.06749451,0.1161203,0.009635141,0.0005184693],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03142075,0.0005457889,0.960897,0.00141324,0.00155362,0.000848198,0.00002231473,0.002128864,0.001170234],"genre_scores_gemma":[0.9549194,0.000003133468,0.04414997,0.00004777697,0.00009658282,0.0004903084,0.00004227878,0.00002361373,0.000226918],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9234987,"threshold_uncertainty_score":0.9999685,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0552265986373305,"score_gpt":0.3262401352911904,"score_spread":0.2710135366538599,"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."}}