FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2553884582> ?p ?o ?g. }

• W2553884582 endingPage "76" @default.
• W2553884582 startingPage "64" @default.
• W2553884582 abstract "The AdaBoost algorithm is a popular ensemble method that combines several weak learners to boost generalization performance. However, conventional AdaBoost.RT algorithms suffer from the limitation that the threshold value must be manually specified rather than chosen through a self-adaptive mechanism, which cannot guarantee a result in an optimal model for general cases. In this paper, we present a generic AdaBoost framework with robust threshold mechanism and structural optimization on regression problems. The error statistics of each weak learner on one given problem dataset is utilized to automate the choice of the optimal cut-off threshold value. In addition, a special single-layer neural network is employed to provide a second opportunity to further adjust the structure and strength the adaption capability of the AdaBoost regression model. Moreover, to consolidate the theoretical foundation of AdaBoost algorithms, we are the first to conduct a rigorous and comprehensive theoretical analysis on the proposed approach. We prove that the general bound on the empirical error with a fraction of training examples is always within a limited soft margin, which indicates that our novel algorithm can avoid over-fitting. We further analyze the bounds on the generalization error directly under probably approximately correct learning. The extensive experimental verifications on the UCI benchmarks have demonstrated that the performance of the proposed method is superior to other state-of-the-art ensemble and single learning algorithms. Furthermore, a real-world indoor positioning application has also revealed that the proposed method has higher positioning accuracy and faster speed." @default.
• W2553884582 created "2016-11-30" @default.
• W2553884582 creator A5003242588 @default.
• W2553884582 creator A5011123518 @default.
• W2553884582 date "2018-01-01" @default.
• W2553884582 modified "2023-10-06" @default.
• W2553884582 title "A Novel AdaBoost Framework With Robust Threshold and Structural Optimization" @default.
• W2553884582 cites W1152476827 @default.
• W2553884582 cites W1566332026 @default.
• W2553884582 cites W1596717185 @default.
• W2553884582 cites W1963989632 @default.
• W2553884582 cites W1981252300 @default.
• W2553884582 cites W1997399503 @default.
• W2553884582 cites W2007146523 @default.
• W2553884582 cites W2015212590 @default.
• W2553884582 cites W2017890988 @default.
• W2553884582 cites W2026131661 @default.
• W2553884582 cites W2051381803 @default.
• W2553884582 cites W2067885219 @default.
• W2553884582 cites W2075170220 @default.
• W2553884582 cites W2096291962 @default.
• W2553884582 cites W2100128988 @default.
• W2553884582 cites W2118207876 @default.
• W2553884582 cites W2121043290 @default.
• W2553884582 cites W2123223828 @default.
• W2553884582 cites W2124193432 @default.
• W2553884582 cites W2125607229 @default.
• W2553884582 cites W2133170554 @default.
• W2553884582 cites W2133176659 @default.
• W2553884582 cites W2135293965 @default.
• W2553884582 cites W2141695047 @default.
• W2553884582 cites W2148446997 @default.
• W2553884582 cites W2153232138 @default.
• W2553884582 cites W2153635508 @default.
• W2553884582 cites W2157595416 @default.
• W2553884582 cites W2169074505 @default.
• W2553884582 cites W2472390602 @default.
• W2553884582 cites W2911964244 @default.
• W2553884582 cites W4212883601 @default.
• W2553884582 cites W4239510810 @default.
• W2553884582 cites W4248437541 @default.
• W2553884582 doi "https://doi.org/10.1109/tcyb.2016.2623900" @default.
• W2553884582 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/27898387" @default.
• W2553884582 hasPublicationYear "2018" @default.
• W2553884582 type Work @default.
• W2553884582 sameAs 2553884582 @default.
• W2553884582 citedByCount "53" @default.
• W2553884582 countsByYear W25538845822017 @default.
• W2553884582 countsByYear W25538845822018 @default.
• W2553884582 countsByYear W25538845822019 @default.
• W2553884582 countsByYear W25538845822020 @default.
• W2553884582 countsByYear W25538845822021 @default.
• W2553884582 countsByYear W25538845822022 @default.
• W2553884582 countsByYear W25538845822023 @default.
• W2553884582 crossrefType "journal-article" @default.
• W2553884582 hasAuthorship W2553884582A5003242588 @default.
• W2553884582 hasAuthorship W2553884582A5011123518 @default.
• W2553884582 hasConcept C11413529 @default.
• W2553884582 hasConcept C119857082 @default.
• W2553884582 hasConcept C12267149 @default.
• W2553884582 hasConcept C134306372 @default.
• W2553884582 hasConcept C141404830 @default.
• W2553884582 hasConcept C149629883 @default.
• W2553884582 hasConcept C153180895 @default.
• W2553884582 hasConcept C154945302 @default.
• W2553884582 hasConcept C177148314 @default.
• W2553884582 hasConcept C178790620 @default.
• W2553884582 hasConcept C185592680 @default.
• W2553884582 hasConcept C33923547 @default.
• W2553884582 hasConcept C41008148 @default.
• W2553884582 hasConcept C45942800 @default.
• W2553884582 hasConcept C50644808 @default.
• W2553884582 hasConcept C774472 @default.
• W2553884582 hasConceptScore W2553884582C11413529 @default.
• W2553884582 hasConceptScore W2553884582C119857082 @default.
• W2553884582 hasConceptScore W2553884582C12267149 @default.
• W2553884582 hasConceptScore W2553884582C134306372 @default.
• W2553884582 hasConceptScore W2553884582C141404830 @default.
• W2553884582 hasConceptScore W2553884582C149629883 @default.
• W2553884582 hasConceptScore W2553884582C153180895 @default.
• W2553884582 hasConceptScore W2553884582C154945302 @default.
• W2553884582 hasConceptScore W2553884582C177148314 @default.
• W2553884582 hasConceptScore W2553884582C178790620 @default.
• W2553884582 hasConceptScore W2553884582C185592680 @default.
• W2553884582 hasConceptScore W2553884582C33923547 @default.
• W2553884582 hasConceptScore W2553884582C41008148 @default.
• W2553884582 hasConceptScore W2553884582C45942800 @default.
• W2553884582 hasConceptScore W2553884582C50644808 @default.
• W2553884582 hasConceptScore W2553884582C774472 @default.
• W2553884582 hasFunder F4320322841 @default.
• W2553884582 hasIssue "1" @default.
• W2553884582 hasLocation W25538845821 @default.
• W2553884582 hasLocation W25538845822 @default.
• W2553884582 hasOpenAccess W2553884582 @default.
• W2553884582 hasPrimaryLocation W25538845821 @default.
• W2553884582 hasRelatedWork W2024223694 @default.
• W2553884582 hasRelatedWork W2061835236 @default.
• W2553884582 hasRelatedWork W3170784702 @default.

• next