Name
Abstract | ||
|---|---|---|
Recently, various dimensionality reduction approaches have been proposed as alternatives to PCA or LDA. These improved approaches do not rely on a linearity assumption, and are hence capable of discovering more complex embeddings within different regions of the data sets. Despite their success on artificial datasets, it is not straightforward to predict which technique is the most appropriate for a given real dataset. In this paper, we empirically evaluate recent techniques on two real audio use cases: musical instrument loops used in music production and sound effects used in sound editing. ISOMAP and t-SNE are being compared to PCA in a visualization problem, where we end up with a two-dimensional view. Various evaluation measures are used: classification performance, as well as trustworthiness/continuity assessing the preservation of neighborhoods. Although PCA and ISOMAP can yield good continuity performance even locally (samples in the original space remain close-by in the low-dimensional one), they fail to preserve the structure of the data well enough to ensure that distinct subgroups remain separate in the visualization. We show that t-SNE presents the best performance, and can even be beneficial as a pre-processing stage for improving classification when the amount of labeled data is low. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/ICME.2013.6607550 | Multimedia and Expo |
Keywords | Field | DocType |
audio signal processing,information retrieval,multimedia computing,music,musical instruments,principal component analysis,ISOMAP,LDA,PCA,artificial datasets,classification performance,complex embeddings,continuity performance,linearity assumption,music production,musical instrument loops,nonlinear dimensionality reduction,real audio use cases,sound editing,sound effects,t-SNE,textural sounds,visualization problem,Dimensionality reduction,audio and music analysis,multimedia information retrieval | Data set,Dimensionality reduction,Computer science,Artificial intelligence,Nonlinear dimensionality reduction,Audio signal processing,Computer vision,Pattern recognition,Visualization,Musical instrument,Machine learning,Principal component analysis,Isomap | Conference |
ISSN | Citations | PageRank |
1945-7871 | 5 | 0.49 |
References | Authors | |
7 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Stéphane Dupont | 1 | 134 | 26.78 |
| Thierry Ravet | 2 | 18 | 4.44 |
| Cécile Picard-Limpens | 3 | 26 | 3.98 |
| Christian Frisson | 4 | 40 | 10.74 |