Application of symbolic data analysis for structural modification assessment

Multivariate analysis of multi-sensor data for assessment of timber structures: Principles and applications

“…Cluster analysis might be also used for classification of decayed samples [25], or species identification [26]. Various algorithms for clustering data were presented by Cury et al [27] during assessing three structural states of a railway bridge. The objective of their research was to use different clustering methods applied to the data in order to classify variety of structural behaviors.…”

Structural modification assessment using supervised learning methods applied to vibration data

“…This combined methodology (SDA + supervised classification methods) has already proven its efficiency when modal parameters are used [11,15]. Now, this paper attempts to answer the following questions: (i) is it possible to classify different structural states using raw data (measured accelerations) only?…”

“…Different types of data can be employed and manipulated in data mining, such as a single quantitative or categorical values, interval-valued data, multi-valued categorical data, and modal multi-valued (histograms) [11,12]. These types of data are generally called ''symbolic data'' and they allow representing the variability and uncertainty present in each variable.…”

Self-organizing map for symbolic data

“…Symbolic variables may present human knowledge, nominal, categorical and synthetic data (see [2,3,19]). Cluster analysis for symbolic data with hierarchical, k-means and fuzzy c-means clustering algorithms had been studied and applied (see [10,20,22,23,38,39]). However, there is less consideration with the SOM learning network for symbolic data.…”

“…IFCM : Class(1) :1, 8,10,11,15,21,22,25,26,31,35,36 Class(2): 4, 6, 7, 9, 13, 16, 18, 19 Class(3) : 3, 5, 12, 20, 23, 24, 32 Class(4) : 2, 14, 17, 27, 28, 29, 33, 34, 37…”