Cross-validation in PCA models with the element-wise k-fold (ekf) algorithm: Practical aspects

Camacho, José; Ferrer, Alberto

doi:10.1016/j.chemolab.2013.12.003

Cited by 51 publications

(69 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This is because PCA models are not prediction models, and a prediction procedure can not be univocally stated. The authors in [31] conclude that the element-wise k-fold (ekf) algorithm is a valid choice for PCA cross-validation when the model is used for missing data imputation purposes, as in the present paper. See the Appendix for a brief explanation of the ekf algorithm.…”

Section: Selection Of the Number Of Latent Variablesmentioning

confidence: 85%

See 1 more Smart Citation

Multivariate Statistical Approach for Anomaly Detection and Lost Data Recovery in Wireless Sensor Networks

Magán-Carrión

Camacho

García-Teodoro

2015

International Journal of Distributed Sensor Networks

Self Cite

View full text Add to dashboard Cite

Data loss due to integrity attacks or malfunction constitutes a principal concern in wireless sensor networks (WSNs). The present paper introduces a novel data loss/modification detection and recovery scheme in this context. Both elements, detection and data recovery, rely on a multivariate statistical analysis approach that exploits spatial density, a common feature in network environments such as WSNs. To evaluate the proposal, we consider WSN scenarios based on temperature sensors, both simulated and real. Furthermore, we consider three different routing algorithms, showing the strong interplay among (a) the routing strategy, (b) the negative effect of data loss on the network performance, and (c) the data recovering capability of the approach. We also introduce a novel data arrangement method to exploit the spatial correlation among the sensors in a more efficient manner. In this data arrangement, we only consider the nearest nodes to a given affected sensor, improving the data recovery performance up to 99%. According to the results, the proposed mechanisms based on multivariate techniques improve the robustness of WSNs against data loss.

show abstract

Section: Selection Of the Number Of Latent Variablesmentioning

confidence: 85%

“…The core of the algorithm performs the recovery of missing values using TSR [31], the output of the algorithm being the matrix of prediction errors E (with elements , in the th row and th column) and the PRESS computed for = 1, . .…”

Section: Appendixmentioning

confidence: 99%

Multivariate Statistical Approach for Anomaly Detection and Lost Data Recovery in Wireless Sensor Networks

Magán-Carrión

Camacho

García-Teodoro

2015

International Journal of Distributed Sensor Networks

Self Cite

View full text Add to dashboard Cite

show abstract

“…As this algorithm, which can be seen as a nonlinear extension of the work in Refs [46,47], does not yield an estimation of σ f , the subsection then introduces an algorithm for optimally estimating σ f when n f is known. This is the first contribution of this article.…”

Section: Objective Function and Algorithm For Kpcamentioning

confidence: 99%

Cross-validatory framework for optimal parameter estimation of KPCA and KPLS models

Krüger

et al. 2017

Chemometrics and Intelligent Laboratory Systems

View full text Add to dashboard Cite

“…These matrices usually are mean The number of components can be determined by crossvalidation (Wold 1978) in such a way that the matrix of the model residuals E does not include a significant predictive component. Note than other criteria than crossvalidation may be more appropriate in Lb-MSPC (Camacho and Ferrer 2014).…”

Section: Latent Variable Models O Principal Component Analysis (Pca)mentioning

confidence: 99%

Fault Diagnosis Tools in Multivariate Statistical Process and Quality Control

Puig¹

View full text Add to dashboard Cite

Agradecimientos AbstractAn accurate fault diagnosis of both, faults sensors and real process faults have become more and more important for process monitoring (minimize downtime, increase safety of plant operation and reduce the manufacturing cost). Quick and correct fault diagnosis is required in order to put back on track our processes or products before safety or quality can be compromised. In the study and comparison of the fault diagnosis methodologies, this thesis distinguishes between two different scenarios, methods for multivariate statistical quality control (MSQC) and methods for latent-based multivariate statistical process control: (Lb-MSPC). In the first part of the thesis the state of the art on fault diagnosis and identification (FDI) is introduced. The second part of the thesis is devoted to the fault diagnosis in multivariate statistical quality control (MSQC). The rationale of the most extended methods for fault diagnosis in supervised scenarios, the requirements for their implementation, their strong points and their drawbacks and relationships are discussed. The performance of the methods is compared using different performance indices in two different process data sets and simulations. New variants and methods to improve the diagnosis performance in MSQC are also proposed. The third part of the thesis is devoted to the fault diagnosis in latent-based multivariate statistical process control (Lb-MSPC). The rationale of the most extended methods for fault diagnosis in supervised Lb-MSPC is described and one of our proposals, the Fingerprints contribution plots (FCP) is introduced. Finally the thesis presents and compare the performance results of these diagnosis methods in Lb-MSPC. The diagnosis results in two process data sets are compared using a new strategy based in the use of the overall sensitivity and specificity. iv v ResumenLa realización de un diagnóstico preciso de los fallos, tanto si se trata de fallos de sensores como si se trata de fallos de procesos, ha llegado a ser algo de vital importancia en la monitorización de procesos (reduce las paradas de planta, incrementa la seguridad de la operación en planta y reduce los costes de producción). Se requieren diagnósticos rápidos y correctos si se quiere poder recuperar los procesos o productos antes de que la seguridad o la calidad de los mismos se pueda ver comprometida. En el estudio de las diferentes metodologías para el diagnóstico de fallos esta tesis distingue dos escenarios diferentes, métodos para el control de estadístico multivariante de la calidad (MSQC) y métodos para el control estadístico de procesos basados en el uso de variables latentes (Lb-MSPC). En la primera parte de esta tesis se introduce el estado del arte sobre el diagnóstico e identificación de fallos (FDI). La segunda parte de la tesis está centrada en el estudio del diagnóstico de fallos en control estadístico multivariante de la calidad. Se describen los fundamentos de los métodos más extendidos para el diagnóstico en escenarios supervisados, sus requerim...

show abstract

Cross-validation in PCA models with the element-wise k-fold (ekf) algorithm: Practical aspects

Cited by 51 publications

References 29 publications

Multivariate Statistical Approach for Anomaly Detection and Lost Data Recovery in Wireless Sensor Networks

Multivariate Statistical Approach for Anomaly Detection and Lost Data Recovery in Wireless Sensor Networks

Cross-validatory framework for optimal parameter estimation of KPCA and KPLS models

Fault Diagnosis Tools in Multivariate Statistical Process and Quality Control

Contact Info

Product

Resources

About