The Relevance Sample-Feature Machine: A Sparse Bayesian Learning Approach to Joint Feature-Sample Selection

Mohsenzadeh, Yalda; Sheikhzadeh, Hamid; Azimi, Reza; Bathaee, Najmehsadat; Kalayeh, Mahdi M.

doi:10.1109/tcyb.2013.2260736

Cited by 52 publications

(20 citation statements)

References 32 publications

(64 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But, few of them consider a joint formulation (Mohsenzadeh et al, 2013). Authors in Mohsenzadeh et al (2013) extend the classic relevance vector machine (RVM) formulation by adding two parameter sets for feature and sample selection in a Bayesian graphical inference model. They consider sparsity in both feature and sample domains, as we do, but instead they solve the problem in a marginal likelihood maximization procedure.…”

Section: Introductionmentioning

confidence: 99%

Joint feature-sample selection and robust diagnosis of Parkinson's disease from MRI data

Adeli

Shi

et al. 2016

NeuroImage

View full text Add to dashboard Cite

Parkinson’s disease (PD) is an overwhelming neurodegenerative disorder caused by deterioration of a neurotransmitter, known as dopamine. Lack of this chemical messenger impairs several brain regions and yields various motor and non-motor symptoms. Incidence of PD is predicted to double in the next two decades, which urges more research to focus on its early diagnosis and treatment. In this paper, we propose an approach to diagnose PD using magnetic resonance imaging (MRI) data. Specifically, we first introduce a joint feature-sample selection (JFSS) method for selecting an optimal subset of samples and features, to learn a reliable diagnosis model. The proposed JFSS model effectively discards poor samples and irrelevant features. As a result, the selected features play an important role in PD characterization, which will help identify the most relevant and critical imaging biomarkers for PD. Then, a robust classification framework is proposed to simultaneously de-noise the selected subset of features and samples, and learn a classification model. Our model can also de-noise testing samples based on the cleaned training data. Unlike many previous works that perform de-noising in an unsupervised manner, we perform supervised de-noising for both training and testing data, thus boosting the diagnostic accuracy. Experimental results on both synthetic and publicly available PD datasets show promising results. To evaluate the proposed method, we use the popular Parkinson’s progression markers initiative (PPMI) database. Our results indicate that the proposed method can differentiate between PD and normal control (NC), and outperforms the competing methods by a relatively large margin. It is noteworthy to mention that our proposed framework can also be used for diagnosis of other brain disorders. To show this, we have also conducted experiments on the widely-used ADNI database. The obtained results indicate that our proposed method can identify the imaging biomarkers and diagnose the disease with favorable accuracies compared to the baseline methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Joint feature-sample selection and robust diagnosis of Parkinson's disease from MRI data

Adeli

Shi

et al. 2016

NeuroImage

View full text Add to dashboard Cite

show abstract

“…However, their work was only focused on unifying frameworks and placed the generalization on broader scale. Mohsenzadeh et al [15] utilized a sparse Bayesian learning approach for feature sample selection. Their proposed relevance sample feature machine (RSFM) is an extension of RVM algorithm.…”

Section: Related Studymentioning

confidence: 99%

Proposing Enhanced Feature Engineering and a Selection Model for Machine Learning Processes

et al. 2018

View full text Add to dashboard Cite

Featured Application: This module can be used independently in any Machine Learning project or can be used in a model that is engineered by boosting and blending of algorithms for better accuracy and fitness.Abstract: Machine Learning (ML) requires a certain number of features (i.e., attributes) to train the model. One of the main challenges is to determine the right number and the type of such features out of the given dataset's attributes. It is not uncommon for the ML process to use dataset of available features without computing the predictive value of each. Such an approach makes the process vulnerable to overfit, predictive errors, bias, and poor generalization. Each feature in the dataset has either a unique predictive value, redundant, or irrelevant value. However, the key to better accuracy and fitting for ML is to identify the optimum set (i.e., grouping) of the right feature set with the finest matching of the feature's value. This paper proposes a novel approach to enhance the Feature Engineering and Selection (eFES) Optimization process in ML. eFES is built using a unique scheme to regulate error bounds and parallelize the addition and removal of a feature during training. eFES also invents local gain (LG) and global gain (GG) functions using 3D visualizing techniques to assist the feature grouping function (FGF). FGF scores and optimizes the participating feature, so the ML process can evolve into deciding which features to accept or reject for improved generalization of the model. To support the proposed model, this paper presents mathematical models, illustrations, algorithms, and experimental results. Miscellaneous datasets are used to validate the model building process in Python, C#, and R languages. Results show the promising state of eFES as compared to the traditional feature selection process.

show abstract

“…7,9,15,22,23 In spite of using kernel method to increase data separability and its reasonable performance in neural decoding, it remains susceptible to the curse of dimensionality. [24][25][26][27] Researchers have applied both supervised and unsupervised dimensionality reduction methods to map neuroimaging data into a lower feature space prior to the classification phase. 28 These techniques diminish the curse of dimensionality and classifier complexity, avoid overfitting, 29,30 and enhance the generalization performance of the classifier.…”

Section: Introductionmentioning

confidence: 99%

Enhancing multivariate pattern analysis for magnetoencephalography through relevant sensor selection

Hatamimajoumerd

Talebpour

Mohsenzadeh

2020

Int J Imaging Syst Tech

Self Cite

View full text Add to dashboard Cite

While the popularity of multivariate pattern classification is growing rapidly in magnetoencephalography (MEG) data analysis, the analysis pipelines used by the neuroscience community are still missing some fundamental machinelearning techniques and principles that would increase their effectiveness.Here, we show that MEG decoding accuracy improves significantly with the addition of feature selection methods to the analysis pipeline. We compare one unsupervised and two supervised feature reduction methods in the current study. Our results show that supervised feature selection methods like statistical dependency and mutual information improve decoding performance and attain higher session-to-session reliability compared to unsupervised dimensionality reduction methods like principal component analysis. Furthermore, we demonstrate that the selected sensors in the data related to a visual task at each time point are consistent with the pattern reflecting the sweep of information in the ventral visual pathway. K E Y W O R D Sdimensionality reduction, feature selection, MEG analysis, MEG sensor selection, multivariate pattern analysis, mutual information, statistical dependency

show abstract

The Relevance Sample-Feature Machine: A Sparse Bayesian Learning Approach to Joint Feature-Sample Selection

Cited by 52 publications

References 32 publications

Joint feature-sample selection and robust diagnosis of Parkinson's disease from MRI data

Joint feature-sample selection and robust diagnosis of Parkinson's disease from MRI data

Proposing Enhanced Feature Engineering and a Selection Model for Machine Learning Processes

Enhancing multivariate pattern analysis for magnetoencephalography through relevant sensor selection

Contact Info

Product

Resources

About