Computationally efficient rule-based classification for continuous streaming data

Advances in hardware and software technologies allow to capture streaming data. The area of Data Stream Mining (DSM) is concerned with the analysis of these vast amounts of data as it is generated in real-time. Data stream classification is one of the most important DSM techniques allowing to classify previously unseen data instances. Different to traditional classifiers for static data, data stream classifiers need to adapt to concept changes (concept drift) in the stream in real-time in order to reflect the most recent concept in the data as accurately as possible. A recent addition to the data stream classifier toolbox is eRules which induces and updates a set of expressive rules that can easily be interpreted by humans. However, like most rule-based data stream classifiers, eRules exhibits a poor computational performance when confronted with continuous attributes. In this work, we propose an approach to deal with continuous data effectively and accurately in rule-based classifiers by using the Gaussian distribution as heuristic for building rule terms on continuous attributes. We show on the example of eRules that incorporating our method for continuous attributes indeed speeds up the real-time rule induction process while maintaining a similar level of accuracy compared with the original eRules classifier. We termed this new version of eRules with our approach G-eRules.

K-Surfer: a KNIME extension for the management and analysis of Human brain MRI FreeSurfer/FSL data

Human brain imaging techniques, such as Magnetic Resonance Imaging (MRI) or Diffusion Tensor Imaging (DTI), have been established as scientific and diagnostic tools and their adoption is growing in popularity. Statistical methods, machine learning and data mining algorithms have successfully been adopted to extract predictive and descriptive models from neuroimage data. However, the knowledge discovery process typically requires also the adoption of pre-processing, post-processing and visualisation techniques in complex data workflows. Currently, a main problem for the integrated preprocessing and mining of MRI data is the lack of comprehensive platforms able to avoid the manual invocation of preprocessing and mining tools, that yields to an error-prone and inefficient process. In this work we present K-Surfer, a novel plug-in of the Konstanz Information Miner (KNIME) workbench, that automatizes the preprocessing of brain images and leverages the mining capabilities of KNIME in an integrated way. K-Surfer supports the importing, filtering, merging and pre-processing of neuroimage data from FreeSurfer, a tool for human brain MRI feature extraction and interpretation. K-Surfer automatizes the steps for importing FreeSurfer data, reducing time costs, eliminating human errors and enabling the design of complex analytics workflow for neuroimage data by leveraging the rich functionalities available in the KNIME workbench.

Phosphorus fertilisation and large legume species affect jarrah forest restoration after bauxite mining

Re-establishing nutrient-cycling is often a key goal of mine-site restoration. This goal can be achieved by applying fertilisers (particularly P) in combination with seeding N-fixing legumes. However, the effect of this strategy on other key restoration goals such as the establishment and growth of non-leguminous species has received little attention. We investigated the effects of P-application rates either singly, or in combination with seeding seven large understorey legume species, on jarrah forest restoration after bauxite mining. Five years after P application and seeding, legume species richness, density and cover were higher in the legume-seeded treatment. However, the increased establishment of legumes did not lead to increased soil N. Increasing P-application rates from 0 to 80 kg P ha−1 did not affect legume species richness, but significantly reduced legume density and increased legume cover: cover was maximal (∼50%) where 80 kg P ha−1 had been applied with large legume seeds. Increasing P-application had no effect on species richness of non-legume species, but increased the density of weeds and native ephemerals. Cover of non-legume species decreased with increasing P-application rates and was lower in plots where large legumes had been seeded compared with non-seeded plots. There was a significant legume × P interaction on weed and ephemeral density: at 80 kg P ha−1 the decline in density of these groups was greatest where legumes were seeded. In addition, the decline in cover for non-legume species with increasing P was greatest when legumes were seeded. Applying 20 kg P ha−1 significantly increased tree growth compared with tree growth in unfertilised plots, but growth was not increased further at 80 kg ha−1 and tree growth was not affected by seeding large legumes. Taken together, these data indicate that 80 kg ha−1 P-fertiliser in combination with (seeding) large legumes maximised vegetation cover at five years but could be suboptimal for re-establishing a jarrah forest community that, like unmined forest, contains a diverse community of slow-growing re-sprouter species. The species richness and cover of non-legume understorey species, especially the resprouters, was highest in plots that received either 0 or 20 kg ha−1 P and where large legumes had not been seeded. Therefore, our findings suggest that moderation of P-fertiliser and legumes could be the best strategy to fulfil the multiple restoration goals of establishing vegetation cover, while at the same time maximising tree growth and species richness of restored forest.

Mining in a changing climate: what scope for forestry-based legacies?

This paper provides an interdisciplinary perspective on mine reclamation in forested areas of Ghana, a country characterised by conflicts between mining and forest conservation. A comparison was made between above ground biomass (AGB) and soil organic carbon (SOC) content from two reclaimed mine sites and adjacent undisturbed forest. Findings suggest that on decadal timescales, reclaimed mine sites contain approximately 40% of the total carbon and 10% the AGB carbon of undisturbed forest. This raises questions regarding the potential for decommissioning mine sites to provide forestry-based legacies. Such a move could deliver a host of benefits, including improving the longevity and success of reclamation, mitigating climate change and delivering corollary enumeration for local communities under carbon trading schemes. A discussion of the antecedents and challenges associated with establishing forest-legacies highlights the risk of neglecting the participation and heterogeneity of legitimate local representatives, which threatens the equity of potential benefits and sustainability of projects. Despite these risks, implementing pilot projects could help to address the lack of transparency and data which currently characterises mine reclamation.