Matches in SemOpenAlex for { <https://semopenalex.org/work/W4251010209> ?p ?o ?g. }
- W4251010209 abstract "Parkinson's disease (PD) is a highly common neurological disease affecting a large population worldwide. Several studies reveal that the degradation of voice is one of its initial symptoms, which is also known as dysarthria. This work explores the correlation between various features in the voice samples observed in PD subjects, using which a novel two-level ensemble-based feature selection method has been proposed. This selected feature set is fed to a custom fine-tuned multilayer perceptron (MLP)-based classifier while using k-fold cross-validation as the resampling strategy. For a robust model training and analysis, three separate benchmark datasets of voice samples were used for the experimentation. Several performance metrics including accuracy, precision, area under the curve (AUC), Matthew's correlation coefficient (MCC), and F1-score were evaluated from the trained model. Results strongly suggest that the proposed feature selection framework helps in identifying an optimal set of features which greatly assist the MLP classifier for a highly accurate identification of PD patients. The proposed model achieves an overall accuracy of 98.3%, 95.1%, and 100% on the three selected datasets. These results are significantly better than those achieved by the non-feature selection-based studies, and even the recently proposed chi-square based feature selection option." @default.
- W4251010209 created "2022-05-12" @default.
- W4251010209 creator A5073136129 @default.
- W4251010209 creator A5073472724 @default.
- W4251010209 creator A5082728145 @default.
- W4251010209 creator A5089000543 @default.
- W4251010209 date "2021-11-01" @default.
- W4251010209 modified "2023-10-16" @default.
- W4251010209 title "An ensemble‐based feature selection framework for early detection of Parkinson's disease based on feature correlation analysis" @default.
- W4251010209 cites W1521663184 @default.
- W4251010209 cites W1534477342 @default.
- W4251010209 cites W1603903339 @default.
- W4251010209 cites W1991307725 @default.
- W4251010209 cites W2006199542 @default.
- W4251010209 cites W2019037755 @default.
- W4251010209 cites W2023294425 @default.
- W4251010209 cites W2031740174 @default.
- W4251010209 cites W2032655680 @default.
- W4251010209 cites W2037760741 @default.
- W4251010209 cites W2069778984 @default.
- W4251010209 cites W2078464941 @default.
- W4251010209 cites W2081070229 @default.
- W4251010209 cites W2090771594 @default.
- W4251010209 cites W2112765817 @default.
- W4251010209 cites W2119387367 @default.
- W4251010209 cites W2134886133 @default.
- W4251010209 cites W2142430469 @default.
- W4251010209 cites W2142769667 @default.
- W4251010209 cites W2143426320 @default.
- W4251010209 cites W2148969844 @default.
- W4251010209 cites W2158143121 @default.
- W4251010209 cites W2165250079 @default.
- W4251010209 cites W2268496675 @default.
- W4251010209 cites W2339549526 @default.
- W4251010209 cites W2344936922 @default.
- W4251010209 cites W2402137650 @default.
- W4251010209 cites W2464977649 @default.
- W4251010209 cites W2496897087 @default.
- W4251010209 cites W2544106530 @default.
- W4251010209 cites W2554665268 @default.
- W4251010209 cites W2557492818 @default.
- W4251010209 cites W2598766133 @default.
- W4251010209 cites W2755642732 @default.
- W4251010209 cites W2788130861 @default.
- W4251010209 cites W2954368032 @default.
- W4251010209 cites W3120747399 @default.
- W4251010209 doi "https://doi.org/10.1002/mma.7835" @default.
- W4251010209 hasPublicationYear "2021" @default.
- W4251010209 type Work @default.
- W4251010209 citedByCount "0" @default.
- W4251010209 crossrefType "journal-article" @default.
- W4251010209 hasAuthorship W4251010209A5073136129 @default.
- W4251010209 hasAuthorship W4251010209A5073472724 @default.
- W4251010209 hasAuthorship W4251010209A5082728145 @default.
- W4251010209 hasAuthorship W4251010209A5089000543 @default.
- W4251010209 hasConcept C117220453 @default.
- W4251010209 hasConcept C119857082 @default.
- W4251010209 hasConcept C138885662 @default.
- W4251010209 hasConcept C148483581 @default.
- W4251010209 hasConcept C150921843 @default.
- W4251010209 hasConcept C153180895 @default.
- W4251010209 hasConcept C154945302 @default.
- W4251010209 hasConcept C179717631 @default.
- W4251010209 hasConcept C2524010 @default.
- W4251010209 hasConcept C27181475 @default.
- W4251010209 hasConcept C2776401178 @default.
- W4251010209 hasConcept C2780092901 @default.
- W4251010209 hasConcept C33923547 @default.
- W4251010209 hasConcept C41008148 @default.
- W4251010209 hasConcept C41895202 @default.
- W4251010209 hasConcept C50644808 @default.
- W4251010209 hasConcept C95623464 @default.
- W4251010209 hasConceptScore W4251010209C117220453 @default.
- W4251010209 hasConceptScore W4251010209C119857082 @default.
- W4251010209 hasConceptScore W4251010209C138885662 @default.
- W4251010209 hasConceptScore W4251010209C148483581 @default.
- W4251010209 hasConceptScore W4251010209C150921843 @default.
- W4251010209 hasConceptScore W4251010209C153180895 @default.
- W4251010209 hasConceptScore W4251010209C154945302 @default.
- W4251010209 hasConceptScore W4251010209C179717631 @default.
- W4251010209 hasConceptScore W4251010209C2524010 @default.
- W4251010209 hasConceptScore W4251010209C27181475 @default.
- W4251010209 hasConceptScore W4251010209C2776401178 @default.
- W4251010209 hasConceptScore W4251010209C2780092901 @default.
- W4251010209 hasConceptScore W4251010209C33923547 @default.
- W4251010209 hasConceptScore W4251010209C41008148 @default.
- W4251010209 hasConceptScore W4251010209C41895202 @default.
- W4251010209 hasConceptScore W4251010209C50644808 @default.
- W4251010209 hasConceptScore W4251010209C95623464 @default.
- W4251010209 hasLocation W42510102091 @default.
- W4251010209 hasOpenAccess W4251010209 @default.
- W4251010209 hasPrimaryLocation W42510102091 @default.
- W4251010209 hasRelatedWork W1990520691 @default.
- W4251010209 hasRelatedWork W2116447301 @default.
- W4251010209 hasRelatedWork W2295032739 @default.
- W4251010209 hasRelatedWork W3016925281 @default.
- W4251010209 hasRelatedWork W3200179079 @default.
- W4251010209 hasRelatedWork W4226324856 @default.
- W4251010209 hasRelatedWork W4249229055 @default.
- W4251010209 hasRelatedWork W4251010209 @default.