Document clustering with K-tree

This paper describes the approach taken to the XML Mining track at INEX 2008 by a group at the Queensland University of Technology. We introduce the K-tree clustering algorithm in an Information Retrieval context by adapting it for document clustering. Many large scale problems exist in document clustering. K-tree scales well with large inputs due to its low complexity. It offers promising results both in terms of efficiency and quality. Document classification was completed using Support Vector Machines.

Modelling the effects of bone fragment contact in fracture healing

The fracture healing process is modulated by the mechanical environment created by imposed loads and motion between the bone fragments. Contact between the fragments obviously results in a significantly different stress and strain environment to a uniform fracture gap containing only soft tissue (e.g. haematoma). The assumption of the latter in existing computational models of the healing process will hence exaggerate the inter-fragmentary strain in many clinically-relevant cases. To address this issue, we introduce the concept of a contact zone that represents a variable degree of contact between cortices by the relative proportions of bone and soft tissue present. This is introduced as an initial condition in a two-dimensional iterative finite element model of a healing tibial fracture, in which material properties are defined by the volume fractions of each tissue present. The algorithm governing the formation of cartilage and bone in the fracture callus uses fuzzy logic rules based on strain energy density resulting from axial compression. The model predicts that increasing the degree of initial bone contact reduces the amount of callus formed (periosteal callus thickness 3.1mm without contact, down to 0.5mm with 10% bone in contact zone). This is consistent with the greater effective stiffness in the contact zone and hence, a smaller inter-fragmentary strain. These results demonstrate that the contact zone strategy reasonably simulates the differences in the healing sequence resulting from the closeness of reduction.

The role of timber plantations in U.K. investment portfolios

Although timber plantations and forests are classified as forms of agricultural production, the ownership of this land classification is not limited to rural producers. Timber plantations and forests are now regarded as a long-term investment with both institutional and absentee owners. While the NCREIF property indices have been the benchmarks for the measurement of the performance of the commercial property market in the UK, for many years the IPD timberland index has recently emerged as the U.K. forest and timberland performance indicator. The IPD Forest index incorporates 126 properties over five regions in the U.K. This paper will utilise the IPD Forestry Index to examine the performance of U.K. timber plantations and forests over the period 1981-2004. In particular, issues to be critically assessed include plantation and forest performance analysis, comparative investment analysis, and the role of plantations and forests in investment portfolios, the risk reduction and portfolio benefits of plantations and forests in mixed-asset portfolios and the strategic investment significance of U.K. timberlands.

The logic of ADHD : a brief review of fallacious reasoning

This paper has two central purposes: the first is to survey some of the more important examples of fallacious argument, and the second is to examine the frequent use of these fallacies in support of the psychological construct: Attention Deficit Hyperactivity Disorder (ADHD). The paper divides 12 familiar fallacies into three different categories—material, psychological and logical—and contends that advocates of ADHD often seem to employ these fallacies to support their position. It is suggested that all researchers, whether into ADHD or otherwise, need to pay much closer attention to the construction of their arguments if they are not to make truth claims unsupported by satisfactory evidence, form or logic.

Australia : LIS education

The topic of library and information science (LIS) education has been under the spotlight in the professional literature in Australia and New Zealand for a number of years. Critical issues of discussion encompass the apparent lack of a core curriculum for the discipline, the perceived gulf between LIS education and LIS practice, and the pressing need for career-long learning and development. One of the central points of debate that emerges repeatedly is the long-standing question about the positioning of the profession: Is LIS a graduate profession of highly skilled individuals valued for their expertise and professionalism or is it a profession of anyone who works in a library, regardless of their qualifications (LIANZA, 2005)? While Australia and New Zealand do not stand alone in this debate – similar issues are echoed in many other countries – there are inevitably some local characteristics which warrant exploration. The discussion presented here highlights the historical background to professional training, the specific professional policies and standards that guide LIS education and some of the challenges facing professional and paraprofessional education, given the changing environment of education in Australia as a whole, with some comparisons made with the New Zealand situation. While all too often library practitioners point the finger at the library educators to ‘right the wrongs’, the authors wish to reinforce the idea that the future of effective and relevant LIS education is a matter for all stakeholders in the profession: practitioners and educators, students and staff, employers and employees, with cohesion potentially offered by the professional body.

Random Indexing K-tree

Random Indexing K-tree is the combination of two algorithms suited for large scale document clustering.

Clustering with random indexing K-tree and XML structure

This paper describes the approach taken to the clustering task at INEX 2009 by a group at the Queensland University of Technology. The Random Indexing (RI) K-tree has been used with a representation that is based on the semantic markup available in the INEX 2009 Wikipedia collection. The RI K-tree is a scalable approach to clustering large document collections. This approach has produced quality clustering when evaluated using two different methodologies.