Is there a preferred classifier for operational thematic mapping?

The importance of properly exploiting a classifier's inherent geometric characteristics when developing a classification methodology is emphasized as a prerequisite to achieving near optimal performance when carrying out thematic mapping. When used properly, it is argued that the long-standing maximum likelihood approach and the more recent support vector machine can perform comparably. Both contain the flexibility to segment the spectral domain in such a manner as to match inherent class separations in the data, as do most reasonable classifiers. The choice of which classifier to use in practice is determined largely by preference and related considerations, such as ease of training, multiclass capabilities, and classification cost. © 1980-2012 IEEE.

Strength correlations of cement-mixed soils using data mapping

Development of comparisons and correlations between the unconfined compressive strength (UCS) and the undrained triaxial compressive strength, qu, is essential for generalising performance and optimising the design of cement-stabilised soils. This paper introduces current work in collecting and collating data from a number of research projects involving both laboratory strength tests performed on identical cement-stabilised soil samples. The research project on cement-stabilised Singapore marine clays at the National University of Singapore has been used as an example to explain the work on comparing and correlating results from both tests by normalising data and constructing contour plots. The effect of variables on strength comparison and correlations was evaluated. The variation in strength correlations was found to be dependent on a number of factors including: soil properties, cement content, curing time and stress, total water/cement ratio, confining stress and strain rate. The results showed that at ~ 100 kPa confining stress, UCS and qu, had similar magnitudes. Correlations between strengths and other design variables are discussed and presented.