A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Statistical Decision-Making by Distance Measures
https://orcid.org/https://orcid.org/0000-0002-4864-2120
Statistical decision-making or feature classification is the final step of a data-driven SHM strategy. This step utilizes features extracted from feature extraction algorithms for damage diagnosis including damage detection, localization, and quantification. In the context of SHM, novelty detection is an unsupervised learning approach to statistical decision-making. Among all novelty detection algorithms, distance-based methods are widely used approaches due to some advantages such as simplicity, computational efficiency, and being non-parametric. This chapter of the book is intended to evaluate some conventional statistical distance measures and then propose several new distance-based novelty detection methods via new distance measures. Depending upon the type of features in terms of being univariate or multivariate and the level of damage diagnosis (detection, localization, and quantification), the proposed methods are divided into univariate and multivariate distance algorithms. In most cases, in order to increase the reliability of damage diagnosis, the process of decision-making in SHM is carried out by threshold limits. This chapter also describes two well-known threshold estimation techniques.
Statistical Decision-Making by Distance Measures
https://orcid.org/https://orcid.org/0000-0002-4864-2120
Statistical decision-making or feature classification is the final step of a data-driven SHM strategy. This step utilizes features extracted from feature extraction algorithms for damage diagnosis including damage detection, localization, and quantification. In the context of SHM, novelty detection is an unsupervised learning approach to statistical decision-making. Among all novelty detection algorithms, distance-based methods are widely used approaches due to some advantages such as simplicity, computational efficiency, and being non-parametric. This chapter of the book is intended to evaluate some conventional statistical distance measures and then propose several new distance-based novelty detection methods via new distance measures. Depending upon the type of features in terms of being univariate or multivariate and the level of damage diagnosis (detection, localization, and quantification), the proposed methods are divided into univariate and multivariate distance algorithms. In most cases, in order to increase the reliability of damage diagnosis, the process of decision-making in SHM is carried out by threshold limits. This chapter also describes two well-known threshold estimation techniques.
Statistical Decision-Making by Distance Measures
https://orcid.org/https://orcid.org/0000-0002-4864-2120
Statistical decision-making or feature classification is the final step of a data-driven SHM strategy. This step utilizes features extracted from feature extraction algorithms for damage diagnosis including damage detection, localization, and quantification. In the context of SHM, novelty detection is an unsupervised learning approach to statistical decision-making. Among all novelty detection algorithms, distance-based methods are widely used approaches due to some advantages such as simplicity, computational efficiency, and being non-parametric. This chapter of the book is intended to evaluate some conventional statistical distance measures and then propose several new distance-based novelty detection methods via new distance measures. Depending upon the type of features in terms of being univariate or multivariate and the level of damage diagnosis (detection, localization, and quantification), the proposed methods are divided into univariate and multivariate distance algorithms. In most cases, in order to increase the reliability of damage diagnosis, the process of decision-making in SHM is carried out by threshold limits. This chapter also describes two well-known threshold estimation techniques.
Statistical Decision-Making by Distance Measures
SpringerBriefs in Applied Sciences
Entezami, Alireza (author)
2021-02-02
21 pages
Article/Chapter (Book)
Electronic Resource
English
Decision Making with Distance Measures and Linguistic Aggregation Operators
| British Library Online Contents | 2010
Fairness measures for decision-making and conflict resolution
| Online Contents | 2019
Fairness measures for decision-making and conflict resolution
| Springer Verlag | 2019
Structural Health Monitoring by Time Series Analysis and Statistical Distance Measures
| UB Braunschweig | 2021