{"id":"W4406583763","doi":"10.1016/j.jnca.2025.104105","title":"Efficient privacy-preserving ML for IoT: Cluster-based split federated learning scheme for non-IID data","year":2025,"lang":"en","type":"article","venue":"Journal of Network and Computer Applications","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université du Québec à Trois-Rivières; École de Technologie Supérieure","funders":"Ministère de la Défense Nationale; Innovation for Defence Excellence and Security","keywords":"Computer science; Scheme (mathematics); Cluster (spacecraft); Internet of Things; Federated learning; Differential privacy; Computer network; Data mining; Computer security; Distributed computing","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":["open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.00119865,0.00019894,0.0003472922,0.0001603179,0.0006636874,0.0004803678,0.0157135,0.0001271394,0.000001423493],"category_scores_gemma":[0.0009569146,0.0001862078,0.00009570004,0.0006938644,0.0000650618,0.0001918095,0.02732544,0.0003649909,0.000001569464],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006061683,"about_ca_system_score_gemma":0.0002466897,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002102853,"about_ca_topic_score_gemma":0.000001727446,"domain_scores_codex":[0.9981385,0.00005421434,0.0006436206,0.0005551863,0.0002042308,0.0004042127],"domain_scores_gemma":[0.9942294,0.001079312,0.0004802124,0.003720159,0.0003982299,0.00009272344],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001154527,0.0003330532,0.001481223,0.0003444756,0.0002685369,0.000002766217,0.00004947382,0.07484615,0.0001829691,0.01181743,0.6730475,0.2375109],"study_design_scores_gemma":[0.0009400402,0.00009840966,0.0003850247,0.000182506,0.00003072858,0.00001212389,0.00000688507,0.8945026,0.00005515477,0.02922659,0.07440562,0.0001542562],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002987099,0.0003340433,0.9775169,0.01773487,0.000289496,0.0009082222,0.00002003373,0.0001541373,0.00005521648],"genre_scores_gemma":[0.04175274,0.00002645298,0.957036,0.0005339471,0.0004275457,0.0001342439,0.00004565125,0.00001543902,0.0000279571],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8196565,"threshold_uncertainty_score":0.989612,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03087237617886379,"score_gpt":0.3047771414575481,"score_spread":0.2739047652786843,"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."}}