SpectralMAP: Approximating Data Manifold With Spectral Decomposition
Dimensionality reduction is widely used to visualize complex high-dimensional data. This study presents a novel method for effective data visualization. Previous methods depend on local distance measurements for data manifold approximation. This leads to unreliable results when a data manifold local...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10070750/ |
| _version_ | 1797853068234063872 |
|---|---|
| author | Koshi Watanabe Keisuke Maeda Takahiro Ogawa Miki Haseyama |
| author_facet | Koshi Watanabe Keisuke Maeda Takahiro Ogawa Miki Haseyama |
| author_sort | Koshi Watanabe |
| collection | DOAJ |
| description | Dimensionality reduction is widely used to visualize complex high-dimensional data. This study presents a novel method for effective data visualization. Previous methods depend on local distance measurements for data manifold approximation. This leads to unreliable results when a data manifold locally oscillates because of some undesirable effects, such as noise effects. In this study, we overcome this limitation by introducing a dual approximation of a data manifold. We roughly approximate a data manifold with a neighborhood graph and prune it with a global filter. This dual scheme results in local oscillation robustness and yields effective visualization with explicit global preservation. We consider a global filter based on principal component analysis frameworks and derive it with the spectral information of the original high-dimensional data. Finally, we experiment with multiple datasets to verify our method, compare its performance to that of state-of-the-art methods, and confirm the effectiveness of our novelty and results. |
| first_indexed | 2024-04-09T19:43:27Z |
| format | Article |
| id | doaj.art-a1cb58812fa04e0b856869201cb7fe08 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-09T19:43:27Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-a1cb58812fa04e0b856869201cb7fe082023-04-03T23:00:31ZengIEEEIEEE Access2169-35362023-01-0111315303154010.1109/ACCESS.2023.325742710070750SpectralMAP: Approximating Data Manifold With Spectral DecompositionKoshi Watanabe0https://orcid.org/0000-0002-0458-802XKeisuke Maeda1https://orcid.org/0000-0001-8039-3462Takahiro Ogawa2https://orcid.org/0000-0001-5332-8112Miki Haseyama3https://orcid.org/0000-0003-1496-1761Graduate School of Information Science and Technology, Hokkaido University, Sapporo, JapanFaculty of Information Science and Technology, Hokkaido University, Sapporo, JapanFaculty of Information Science and Technology, Hokkaido University, Sapporo, JapanFaculty of Information Science and Technology, Hokkaido University, Sapporo, JapanDimensionality reduction is widely used to visualize complex high-dimensional data. This study presents a novel method for effective data visualization. Previous methods depend on local distance measurements for data manifold approximation. This leads to unreliable results when a data manifold locally oscillates because of some undesirable effects, such as noise effects. In this study, we overcome this limitation by introducing a dual approximation of a data manifold. We roughly approximate a data manifold with a neighborhood graph and prune it with a global filter. This dual scheme results in local oscillation robustness and yields effective visualization with explicit global preservation. We consider a global filter based on principal component analysis frameworks and derive it with the spectral information of the original high-dimensional data. Finally, we experiment with multiple datasets to verify our method, compare its performance to that of state-of-the-art methods, and confirm the effectiveness of our novelty and results.https://ieeexplore.ieee.org/document/10070750/Data visualizationdimensionality reductionspectral-based filtering |
| spellingShingle | Koshi Watanabe Keisuke Maeda Takahiro Ogawa Miki Haseyama SpectralMAP: Approximating Data Manifold With Spectral Decomposition IEEE Access Data visualization dimensionality reduction spectral-based filtering |
| title | SpectralMAP: Approximating Data Manifold With Spectral Decomposition |
| title_full | SpectralMAP: Approximating Data Manifold With Spectral Decomposition |
| title_fullStr | SpectralMAP: Approximating Data Manifold With Spectral Decomposition |
| title_full_unstemmed | SpectralMAP: Approximating Data Manifold With Spectral Decomposition |
| title_short | SpectralMAP: Approximating Data Manifold With Spectral Decomposition |
| title_sort | spectralmap approximating data manifold with spectral decomposition |
| topic | Data visualization dimensionality reduction spectral-based filtering |
| url | https://ieeexplore.ieee.org/document/10070750/ |
| work_keys_str_mv | AT koshiwatanabe spectralmapapproximatingdatamanifoldwithspectraldecomposition AT keisukemaeda spectralmapapproximatingdatamanifoldwithspectraldecomposition AT takahiroogawa spectralmapapproximatingdatamanifoldwithspectraldecomposition AT mikihaseyama spectralmapapproximatingdatamanifoldwithspectraldecomposition |