Soil mix technology for integrated remediation and ground improvement: From laboratory work to field trials

Relatively new in the UK, soil mix technology applied to the in-situ remediation of contaminated land involves the use of mixing tools and additives to construct permeable reactive in-ground barriers and low-permeability containment walls and for hot-spot soil treatment by stabilisation/ solidification. It is a cost effective and versatile approach with numerous environmental advantages. Further commercial advantages can be realised by combining this with ground improvement through the development of a single integrated soil mix technology system which is the core objective of Project SMiRT (Soil Mix Remediation Technology). This is a large UK-based R&D project involving academia-industry collaboration with a number of tasks including equipment development, laboratory treatability studies, field trials, stakeholder consultation and dissemination activities. This paper presents aspects of project SMiRT relating to the laboratory treatability study work leading to the design of the field trials. © 2012 American Society of Civil Engineers.

Randomized nonlinear component analysis

Copyright © (2014) by the International Machine Learning Society (IMLS) All rights reserved. Classical methods such as Principal Component Analysis (PCA) and Canonical Correlation Analysis (CCA) are ubiquitous in statistics. However, these techniques are only able to reveal linear re-lationships in data. Although nonlinear variants of PCA and CCA have been proposed, these are computationally prohibitive in the large scale. In a separate strand of recent research, randomized methods have been proposed to construct features that help reveal nonlinear patterns in data. For basic tasks such as regression or classification, random features exhibit little or no loss in performance, while achieving drastic savings in computational requirements. In this paper we leverage randomness to design scalable new variants of nonlinear PCA and CCA; our ideas extend to key multivariate analysis tools such as spectral clustering or LDA. We demonstrate our algorithms through experiments on real- world data, on which we compare against the state-of-the-art. A simple R implementation of the presented algorithms is provided.