Selection of relevant features for audio classification tasks

2011 ◽  
Author(s):  
Μαρία Μαρκάκη
Keyword(s):  
Audio Classification ◽  
Classification Tasks ◽  
Selection Of

scholarly journals Selection of Suitable Machine Learning Algorithms for Classification Tasks in Reverse Logistics

Procedia CIRP ◽  
2021 ◽  
Vol 96 ◽  
pp. 272-277
Author(s):  
Hannah Lickert ◽  
Aleksandra Wewer ◽  
Sören Dittmann ◽  
Pinar Bilge ◽  
Franz Dietrich
Keyword(s):  
Machine Learning ◽  
Reverse Logistics ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Classification Tasks ◽  
Selection Of

scholarly journals Evolutionary Multitasking-Based Multiobjective Optimization Algorithm for Channel Selection in Hybrid Brain Computer Interfacing Systems

2021 ◽  
Vol 15 ◽  
Author(s):  
Tianyu Liu ◽  
Zhixiong Xu ◽  
Lei Cao ◽  
Guowei Tan
Keyword(s):  
Multiobjective Optimization ◽  
Real Life ◽  
Channel Selection ◽  
Research Field ◽  
Neural Engineering ◽  
Computer Interfaces ◽  
Classification Tasks ◽  
Optimization Methodology ◽  
Brain Computer Interfacing ◽  
Selection Of

Hybrid-modality brain-computer Interfaces (BCIs), which combine motor imagery (MI) bio-signals and steady-state visual evoked potentials (SSVEPs), has attracted wide attention in the research field of neural engineering. The number of channels should be as small as possible for real-life applications. However, most of recent works about channel selection only focus on either the performance of classification task or the effectiveness of device control. Few works conduct channel selection for MI and SSVEP classification tasks simultaneously. In this paper, a multitasking-based multiobjective evolutionary algorithm (EMMOA) was proposed to select appropriate channels for these two classification tasks at the same time. Moreover, a two-stage framework was introduced to balance the number of selected channels and the classification accuracy in the proposed algorithm. The experimental results verified the feasibility of multiobjective optimization methodology for channel selection of hybrid BCI tasks.

scholarly journals Data Augmentation and Feature Selection for Automatic Model Recommendation in Computational Physics

2021 ◽  
Vol 26 (1) ◽  
pp. 17
Author(s):  
Thomas Daniel ◽  
Fabien Casenave ◽  
Nissrine Akkari ◽  
David Ryckelynck
Keyword(s):  
Feature Selection ◽  
Data Augmentation ◽  
Classification Problem ◽  
Computational Physics ◽  
Training Data ◽  
Multilayer Perceptrons ◽  
Augmentation Techniques ◽  
Physical Fields ◽  
Classification Tasks ◽  
Selection Of

Classification algorithms have recently found applications in computational physics for the selection of numerical methods or models adapted to the environment and the state of the physical system. For such classification tasks, labeled training data come from numerical simulations and generally correspond to physical fields discretized on a mesh. Three challenging difficulties arise: the lack of training data, their high dimensionality, and the non-applicability of common data augmentation techniques to physics data. This article introduces two algorithms to address these issues: one for dimensionality reduction via feature selection, and one for data augmentation. These algorithms are combined with a wide variety of classifiers for their evaluation. When combined with a stacking ensemble made of six multilayer perceptrons and a ridge logistic regression, they enable reaching an accuracy of 90% on our classification problem for nonlinear structural mechanics.

Benchmarking Functional Link Expansions for Audio Classification Tasks

2016 ◽  
pp. 133-141 ◽  
Author(s):  
Simone Scardapane ◽  
Danilo Comminiello ◽  
Michele Scarpiniti ◽  
Raffaele Parisi ◽  
Aurelio Uncini
Keyword(s):  
Audio Classification ◽  
Functional Link ◽  
Classification Tasks

Fault Diagnosis of Power Quality

2011 ◽  
Vol 128-129 ◽  
pp. 870-873
Author(s):  
Yang Lan Ou
Keyword(s):  
Wavelet Transform ◽  
Fault Diagnosis ◽  
Power Quality ◽  
Performance Comparison ◽  
Rule Based ◽  
Event Classification ◽  
Minimum Error ◽  
Fuzzy Classifiers ◽  
Classification Tasks ◽  
Selection Of

The proposed method in this paper indicates two issues, selection of discriminative features and classifies event classes with minimum error. Wavelets features (WF) of power quality (PQ) events are extracted using wavelet transform (WT) and fuzzy classifiers classify events using these features. The captured signals are often corrupted by noise; the non-linear and non-stationary behaviors of PQ events make the detection and classification tasks more cumbersome. Performance comparison of the proposed method is made with three other fuzzy classifiers using different wavelets and superiority is verified. In the proposed approach of event classification, fuzzy product aggregation reasoning rule based method has been used.

scholarly journals Over-Parameterization and Generalization in Audio Classification

2021 ◽  
Author(s):  
Khaled Koutini ◽  
Hamid Eghbal-zadeh ◽  
Florian Henkel ◽  
Jan Schlüter ◽  
Gerhard Widmer
Keyword(s):  
Neural Networks ◽  
Language Processing ◽  
Recording Device ◽  
Classification Models ◽  
Audio Classification ◽  
Scene Classification ◽  
Machine Listening ◽  
Substantial Problem ◽  
Classification Tasks ◽  
The Relationship

Convolutional Neural Networks (CNNs) have been dominating classification tasks in various domains, such as machine vision, machine listening, and natural language processing. In machine listening, while generally exhibiting very good generalization capabilities, CNNs are sensitive to the specific audio recording device used, which has been recognized as a substantial problem in the acoustic scene classification (DCASE) community. In this study, we investigate the relationship between over-parameterization of acoustic scene classification models, and their resulting generalization abilities. Our results indicate that increasing width improves generalization to unseen devices, even without an increase in the number of parameters.

Combining visual and acoustic features for audio classification tasks

2017 ◽  
Vol 88 ◽  
pp. 49-56 ◽  
Author(s):  
L. Nanni ◽  
Y.M.G. Costa ◽  
D.R. Lucio ◽  
C.N. Silla ◽  
S. Brahnam
Keyword(s):  
Audio Classification ◽  
Acoustic Features ◽  
Classification Tasks

Parameters Selection of LLE Algorithm for Classification Tasks

2014 ◽  
Vol 1037 ◽  
pp. 422-427 ◽  
Author(s):  
Feng Hu ◽  
Chuan Tong Wang ◽  
Yu Chuan Wu ◽  
Liang Zhi Fan
Keyword(s):  
Real Life ◽  
Real Data ◽  
Locally Linear Embedding ◽  
Data Sets ◽  
Theoretical Argument ◽  
Class Separability ◽  
Parameters Selection ◽  
Classification Tasks ◽  
Neighborhood Topology ◽  
Selection Of

The crux in the locally linear embedding algorithm or LLE is the selection of embedding dimensionality and neighborhood size. A method of parameters selection based on the normalized cut criterion or Ncut for classification tasks is proposed. Differing from current techniques based on the neighborhood topology preservation criterion, the proposed method capitalizes on class separability of embedding result. By taking it into consideration, the intrinsic capability of LLE can be more faithfully reflected, and hence more rational features for classification in real-life applications can be offered. The theoretical argument is supported by experimental results from synthetic and real data sets.

Cognitive gadgets and cognitive priors

2019 ◽  
Vol 42 ◽  
Author(s):  
Gian Domenico Iannetti ◽  
Giorgio Vallortigara
Keyword(s):  
Cognitive Neuroscience ◽  
Selection Of

Abstract Some of the foundations of Heyes’ radical reasoning seem to be based on a fractional selection of available evidence. Using an ethological perspective, we argue against Heyes’ rapid dismissal of innate cognitive instincts. Heyes’ use of fMRI studies of literacy to claim that culture assembles pieces of mental technology seems an example of incorrect reverse inferences and overlap theories pervasive in cognitive neuroscience.

Sign in / Sign up

Export Citation Format

Share Document