Nonlinear Feature Extraction Through Manifold Learning in an Electronic Tongue Classification Task
A nonlinear feature extraction-based approach using manifold learning algorithms is developed in order to improve the classification accuracy in an electronic tongue sensor array. The developed signal processing methodology is composed of four stages: data unfolding, scaling, feature extraction, and...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/17/4834 |
| _version_ | 1827707456400130048 |
|---|---|
| author | Jersson X. Leon-Medina Maribel Anaya Francesc Pozo Diego Tibaduiza |
| author_facet | Jersson X. Leon-Medina Maribel Anaya Francesc Pozo Diego Tibaduiza |
| author_sort | Jersson X. Leon-Medina |
| collection | DOAJ |
| description | A nonlinear feature extraction-based approach using manifold learning algorithms is developed in order to improve the classification accuracy in an electronic tongue sensor array. The developed signal processing methodology is composed of four stages: data unfolding, scaling, feature extraction, and classification. This study aims to compare seven manifold learning algorithms: Isomap, Laplacian Eigenmaps, Locally Linear Embedding (LLE), modified LLE, Hessian LLE, Local Tangent Space Alignment (LTSA), and <i>t</i>-Distributed Stochastic Neighbor Embedding (<i>t</i>-SNE) to find the best classification accuracy in a multifrequency large-amplitude pulse voltammetry electronic tongue. A sensitivity study of the parameters of each manifold learning algorithm is also included. A data set of seven different aqueous matrices is used to validate the proposed data processing methodology. A leave-one-out cross validation was employed in 63 samples. The best accuracy (<inline-formula><math display="inline"><semantics><mrow><mn>96.83</mn><mo>%</mo></mrow></semantics></math></inline-formula>) was obtained when the methodology uses Mean-Centered Group Scaling (MCGS) for data normalization, the <i>t</i>-SNE algorithm for feature extraction, and <i>k</i>-nearest neighbors (<i>k</i>NN) as classifier. |
| first_indexed | 2024-03-10T16:46:54Z |
| format | Article |
| id | doaj.art-69c6ae7c4c964e82a0db160d04f7253e |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T16:46:54Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-69c6ae7c4c964e82a0db160d04f7253e2023-11-20T11:31:28ZengMDPI AGSensors1424-82202020-08-012017483410.3390/s20174834Nonlinear Feature Extraction Through Manifold Learning in an Electronic Tongue Classification TaskJersson X. Leon-Medina0Maribel Anaya1Francesc Pozo2Diego Tibaduiza3Departamento de Ingeniería Mecánica y Mecatrónica, Universidad Nacional de Colombia, Cra 45 No. 26-85, Bogotá 111321, ColombiaMEM (Modelling-Electronics and Monitoring Research Group), Faculty of Electronics Engineering, Universidad Santo Tomás, Bogotá 110231, ColombiaControl, Modeling, Identification and Applications (CoDAlab), Department of Mathematics, Escola d’Enginyeria de Barcelona Est (EEBE), Universitat Politècnica de Catalunya (UPC), Campus Diagonal-Besòs (CDB), Eduard Maristany, 16, 08019 Barcelona, SpainDepartamento de Ingeniería Eléctrica y Electrónica, Universidad Nacional de Colombia, Cra 45 No. 26-85, Bogotá 111321, ColombiaA nonlinear feature extraction-based approach using manifold learning algorithms is developed in order to improve the classification accuracy in an electronic tongue sensor array. The developed signal processing methodology is composed of four stages: data unfolding, scaling, feature extraction, and classification. This study aims to compare seven manifold learning algorithms: Isomap, Laplacian Eigenmaps, Locally Linear Embedding (LLE), modified LLE, Hessian LLE, Local Tangent Space Alignment (LTSA), and <i>t</i>-Distributed Stochastic Neighbor Embedding (<i>t</i>-SNE) to find the best classification accuracy in a multifrequency large-amplitude pulse voltammetry electronic tongue. A sensitivity study of the parameters of each manifold learning algorithm is also included. A data set of seven different aqueous matrices is used to validate the proposed data processing methodology. A leave-one-out cross validation was employed in 63 samples. The best accuracy (<inline-formula><math display="inline"><semantics><mrow><mn>96.83</mn><mo>%</mo></mrow></semantics></math></inline-formula>) was obtained when the methodology uses Mean-Centered Group Scaling (MCGS) for data normalization, the <i>t</i>-SNE algorithm for feature extraction, and <i>k</i>-nearest neighbors (<i>k</i>NN) as classifier.https://www.mdpi.com/1424-8220/20/17/4834manifold learningfeature extractionclassificationelectronic tonguemachine learningt-SNE |
| spellingShingle | Jersson X. Leon-Medina Maribel Anaya Francesc Pozo Diego Tibaduiza Nonlinear Feature Extraction Through Manifold Learning in an Electronic Tongue Classification Task Sensors manifold learning feature extraction classification electronic tongue machine learning t-SNE |
| title | Nonlinear Feature Extraction Through Manifold Learning in an Electronic Tongue Classification Task |
| title_full | Nonlinear Feature Extraction Through Manifold Learning in an Electronic Tongue Classification Task |
| title_fullStr | Nonlinear Feature Extraction Through Manifold Learning in an Electronic Tongue Classification Task |
| title_full_unstemmed | Nonlinear Feature Extraction Through Manifold Learning in an Electronic Tongue Classification Task |
| title_short | Nonlinear Feature Extraction Through Manifold Learning in an Electronic Tongue Classification Task |
| title_sort | nonlinear feature extraction through manifold learning in an electronic tongue classification task |
| topic | manifold learning feature extraction classification electronic tongue machine learning t-SNE |
| url | https://www.mdpi.com/1424-8220/20/17/4834 |
| work_keys_str_mv | AT jerssonxleonmedina nonlinearfeatureextractionthroughmanifoldlearninginanelectronictongueclassificationtask AT maribelanaya nonlinearfeatureextractionthroughmanifoldlearninginanelectronictongueclassificationtask AT francescpozo nonlinearfeatureextractionthroughmanifoldlearninginanelectronictongueclassificationtask AT diegotibaduiza nonlinearfeatureextractionthroughmanifoldlearninginanelectronictongueclassificationtask |