Extracting common subtrees from decision trees

This paper explores an efficient technique for the extraction of common subtrees in decision trees. The method is based on a Suffix Tree string matching process and the algorithm is applied to the problem of finding common decision rules in path planning.

Video indexing and similarity retrieval by largest common subgraph detection using decision trees

While the largest common subgraph (LCSG) between a query and a database of models can provide an elegant and intuitive measure of similarity for many applications, it is computationally expensive to compute. Recently developed algorithms for subgraph isomorphism detection take advantage of prior knowledge of a database of models to improve the speed of on-line matching. This paper presents a new algorithm based on similar principles to solve the largest common subgraph problem. The new algorithm significantly reduces the computational complexity of detection of the LCSG between a known database of models, and a query given on-line.

Bayesian networks and decision trees in the diagnosis of female urinary incontinence

This study compares the effectiveness of Bayesian networks versus Decision Trees in modeling the Integral Theory of Female Urinary Incontinence diagnostic algorithm. Bayesian networks and Decision Trees were developed and trained using data from 58 adult women presenting with urinary incontinence symptoms. A Bayesian Network was developed in collaboration with an expert specialist who regularly utilizes a non-automated diagnostic algorithm in clinical practice. The original Bayesian network was later refined using a more connected approach. Diagnoses determined from all automated approaches were compared with the diagnoses of a single human expert. In most cases, Bayesian networks were found to be at least as accurate as the Decision Tree approach. The refined Connected Bayesian Network was found to be more accurate than the Original Bayesian Network accurately discriminated between diagnoses despite the small sample size. In contrast, the Connected and Decision Tree approaches were less able to discriminate between diagnoses. The Original Bayesian Network was found to provide an excellent basis for graphically communicating the correlation between symptoms and laxity defects in a given anatomical zone. Performance measures in both networks indicate that Bayesian networks could provide a potentially useful tool in the management of female pelvic floor dysfunction. Before the technique can be utilized in practice, well-established learning algorithms should be applied to improve network structure. A larger training data set should also improve network accuracy, sensitivity, and specificity.

A comparative analysis of decision trees vis-à-vis other computational data mining techniques in automotive insurance fraud detection

The development and application of computational data mining techniques in financial fraud detection and business failure prediction has become a popular cross-disciplinary research area in recent times involving financial economists, forensic accountants and computational modellers. Some of the computational techniques popularly used in the context of - financial fraud detection and business failure prediction can also be effectively applied in the detection of fraudulent insurance claims and therefore, can be of immense practical value to the insurance industry. We provide a comparative analysis of prediction performance of a battery of data mining techniques using real-life automotive insurance fraud data. While the data we have used in our paper is US-based, the computational techniques we have tested can be adapted and generally applied to detect similar insurance frauds in other countries as well where an organized automotive insurance industry exists.

Two trees: Australian artists' books to Afghanistan and back

Two Trees documents an artists' books collaboration between 14 Australian women artists and women in Afghanistan who were learning literacy. The collaboration resulted in 36 concertina books of images, withDari and Pashto hand written texts.The books are held in the State Library of Queensland's artists' books collection. Two Trees includes includes an introduction by Barbara Kameniar, and texts by Malalai Joya, Gali Weiss, Matthias Thomczak and Latifa Ahmady.

Urban to forest gradients: suitability for hollow bearing trees and implications for obligate hollow nesters

 Resource availability is a limiting factor influencing the distribution and composition of faunal communities. Globally, hollow bearing trees are a resource required by wildlife at all trophic levels, and are used for a diverse range of ecological functions. In the northern hemisphere avian species act as primary hollow excavators, whereas the southern hemisphere must rely on complex interactions between stochastic events, and eventual decay. Hollow formation is therefore a slow process in the southern hemisphere. In contrast, hollow loss is quite rapid and influenced greatly by anthropogenic impacts.To identify the ecological characteristics driving hollows over an urban to forest gradient as a resource for the powerful owl (Ninox strenua) and its prey we used presence-only modelling. The potential for an area to support tree hollows suitable for powerful owls and their prey was linked to the density of ephemeral rivers, land cover, tree cover and distance from riparian vegetation. The potential for large hollows throughout the landscape, suitable for the powerful owl, was also influenced by density of permanent rivers. Potential habitat for tree hollows, capable of supporting powerful owls and their prey was greatest in forested environments, declining with increased urbanization. However the urban region still supported some smaller tree hollows suitable for arboreal marsupials. Managing for urban dwelling species, is not as simple as retaining old hollow producing trees or providing alternate nesting structures. We also need to mitigate increased mortality associated with built environments (e.g. electrocution, collisions).

Is there more soil carbon under nitrogen-fixing trees than under non-nitrogen-fixing trees in mixed-species restoration plantings?

Afforestation of agricultural land provides an important opportunity to mitigate climate change by storing carbon (C) in both plant biomass and the soil. Here we present results of a study in which we sought to determine whether soil under nitrogen(N)-fixing trees contained more C than soil under non-N-fixing trees in mixed-species plantings, and thus if inclusion of N-fixers is beneficial in terms of increasing soil C sequestration. Soils were sampled directly beneath N-fixing and non-N-fixing tree species in riparian and upland mixed-species plantings in southeastern Australia. Soil C and N contents were assessed at both the landscape and individual planting scales. At the landscape scale, there were higher levels of soil C and N under N-fixing trees compared with non-N-fixing trees. At the individual planting scale, the patterns were less clear with both large increases and decreases occurring across the range of sites. The results presented here indicate that the inclusion of N-fixers may help to increase soil C, and N, but that the response may be site- and species-specific. © 2014 Elsevier B.V.

N-fixing trees in restoration plantings: effects on nitrogen supply and soil microbial communities

Mixed-species restoration tree plantings are being established increasingly, contributing to mitigate climate change and restore ecosystems. Including nitrogen (N)-fixing tree species may increase carbon (C) sequestration in mixed-species plantings, as these species may substantially increase soil C beneath them. We need to better understand the role of N-fixers in mixed-species plantings to potentially maximize soil C sequestration in these systems. Here, we present a field-based study that asked two specific questions related to the inclusion of N-fixing trees in a mixed-species planting: 1) Do non-N-fixing trees have access to N derived from fixation of atmospheric N2 by neighbouring N-fixing trees? 2) Do soil microbial communities differ under N-fixing trees and non-N-fixing trees in a mixed-species restoration planting? We sampled leaves from the crowns, and litter and soils beneath the crowns of two N-fixing and two non-N-fixing tree species that dominated the planting. Using the 15N natural abundance method, we found indications that fixed atmospheric N was utilized by the non-N-fixing trees, most likely through tight root connections or organic forms of N from the litter layer, rather than through the decomposition of N-fixers litter. While the two N-fixing tree species that were studied appeared to fix atmospheric N, they were substantially different in terms of C and N addition to the soil, as well as microbial community composition beneath them. This shows that the effect of N-fixing tree species on soil carbon sequestration is species-specific, cannot be generalized and requires planting trails to determine if there will be benefits to carbon sequestration. © 2014 Elsevier Ltd.

Failing to see the woods for the trees: in search of critical creative research

Not all examples of creativity can be explained by the functionalist view that creativity is inherent in an artefact and comes about due to a “creative” individual’s efforts within a conducive environment. I build a case for creativity research which overturns assumptions inherent in functionalist research and conceptualises the phenomenon as a context specific, social construction. However, reviewing the limited critical research, I find that there is inconsistency in the theoretical models and the empirical work and ongoing privileging of the individual. It is unsurprising therefore, that, this research has been criticised for causing fragmentation within the field. I argue, to build knowledge, critical creativity research needs to adopt a systemic perspective, focus on social processes and consider collectives.