Convex hulls as an hypothesis language bias

Classification learning is dominated by systems which induce large numbers of small axis-orthogonal decision surfaces which biases such systems towards particular hypothesis types. However, there is reason to believe that many domains have underlying concepts which do not involve axis orthogonal surfaces. Further, the multiplicity of small decision regions mitigates against any holistic appreciation of the theories produced by these systems, notwithstanding the fact that many of the small regions are individually comprehensible. We propose the use of less strongly biased hypothesis languages which might be expected to model' concepts using a number of structures close to the number of actual structures in the domain. An instantiation of such a language, a convex hull based classifier, CHI, has been implemented to investigate modeling concepts as a small number of large geometric structures in n-dimensional space. A comparison of the number of regions induced is made against other well-known systems on a representative selection of largely or wholly continuous valued machine learning tasks. The convex hull system is shown to produce a number of induced regions about an order of magnitude less than well-known systems and very close to the number of actual concepts. This representation, as convex hulls, allows the possibility of extraction of higher level mathematical descriptions of the induced concepts, using the techniques of computational geometry.

Convex hulls in concept induction

Classification learning is dominated by systems which induce large numbers of small axis-orthogonal decision surfaces. This strongly biases such systems towards particular hypothesis types but there is reason believe that many domains have underlying concepts which do not involve axis orthogonal surfaces. Further, the multiplicity of small decision regions mitigates against any holistic appreciation of the theories produced by these systems, notwithstanding the fact that many of the small regions are individually comprehensible. This thesis investigates modeling concepts as large geometric structures in n-dimensional space. Convex hulls are a superset of the set of axis orthogonal hyperrectangles into which axis orthogonal systems partition the instance space. In consequence, there is reason to believe that convex hulls might provide a more flexible and general learning bias than axis orthogonal regions. The formation of convex hulls around a group of points of the same class is shown to be a usable generalisation and is more general than generalisations produced by axis-orthogonal based classifiers, without constructive induction, like decision trees, decision lists and rules. The use of a small number of large hulls as a concept representation is shown to provide classification performance which can be better than that of classifiers which use a large number of small fragmentary regions for each concept. A convex hull based classifier, CH1, has been implemented and tested. CH1 can handle categorical and continuous data. Algorithms for two basic generalisation operations on hulls, inflation and facet deletion, are presented. The two operations are shown to improve the accuracy of the classifier and provide moderate classification accuracy over a representative selection of typical, largely or wholly continuous valued machine learning tasks. The classifier exhibits superior performance to well-known axis-orthogonal-based classifiers when presented with domains where the underlying decision surfaces are not axis parallel. The strengths and weaknesses of the system are identified. One particular advantage is the ability of the system to model domains with approximately the same number of structures as there are underlying concepts. This leads to the possibility of extraction of higher level mathematical descriptions of the induced concepts, using the techniques of computational geometry, which is not possible from a multiplicity of small regions.

Planing a thesis: ways, means and tactics for research through design

"Design is messy and over analysis will stop the flow ... " was the parting line from a colleague after a conversation on design methods. Fo1- many, 'method' is associated with a grim period in a1-chitecture where science held sway. I certainly have no intention of resu1-recting the dogma of that period, but in equal measure I agree with Till (20 12) that painting the design researcher as a heroic genius, marginalises the agency of design research. Despite the various arguments surrounding the value of design as research such as those developed by Downton (2003), van Schaik (2008), or Fraser (2013), there are only a few contempo1-ary resources Uomaka,2008) (Muckenheim and Demel, 2012), which at best provide partial references for students planning a design thesis. If the research emphasis is on designing, then critical knowledge and skills on the ways and means (methods) and tactics for their deployment (methodology) is essential.

Promoting community capacity through university-community engagement: the Deakin University / Department of Human Services (Barwon South West) Partnership

The formal Partnership between Deakin University and the Victorian Department of Human Services (Barwon-South Western Region), based in Geelong, aims to bring together the knowledge, experience and resources of the Department and Deakin University for the benefit of the people living in that region, as well as for the mutual benefit of both organisations. A recent review process featured stakeholder interviews and focus groups. A special workshop on university-community engagement was also held for interested stakeholders in late 2006. This was facilitated by Prof. Judith Ramaley, President of Winona State University, during her visit to Deakin University as a Fulbright Visiting Senior Specialist. Visioning and strategic planing have continued throughout 2007.

This paper will describe the efforts and achievements of the Partnership through the complementary lenses of Healthy Cities, health-promoting universities and community capacity. This framework will be used to describe how the Partnership coordinators have used Prof Ramaley’s insights to establish a draft Business Plan that espouses a more mature form of collaboration and embraces shared, transformative goals. The paper will describe how the notion of community capacity is being used to evaluate the Partnership’s overall contribution to community engagement.

The use of analytic hierarchy process to incorporate stakeholder preferences into regional forest planning

Forest management decisions are often characterised by complexity, irreversibility and uncertainty. Much of the complexity arises from the multiple-use nature of forest goods and services, difficulty in monetary valuation of ecological services and the involvement of numerous stakeholders. Under these circumstances, conventional methods such as cost-benefit analysis are ill-suited to evaluate forest decisions. The Analytic Hierarchy Process (AHP), can be useful in regional forest planing as it can accommodate conflictual, multidimensional, incommensurable and incomparable set of objectives. The objective of this paper is to examine the scope and feasibility of the AHP in incorporating stakeholder preferences into regional forest planning. The Australian Regional Forest Agreement Programme is taken as an illustrative case for the analysis. The results show that the AHP can formalize public participation in decision making and increase the transparency and the credibility of the process.

Characterization of the moisture absorption and thermal ageing behaviour of polymeric composite systems using Raman spectroscopy

Advanced polymeric materials and their respective composites are fast becoming one of the world's most frequently used engineering materials. They find application in the manufacture of e.g. boat hulls, high performance motor vehicles, aircraft components and sports goods. Their high specific strength and specific stiffness give them the edge in applications where weight savings are critical, but their long-term durability is often questioned. These materials are susceptible to environmental conditions such as temperature and humidity. There is also a lack of relevant data, due to the long time-scales required for testing. In this study, the Raman technique has been used to monitor the degradation of two composite systems, namely: a rubber toughened vinylester material used in the marine industry and a high temperature bismaleimide/carbon fibre aerospace composite. Preliminary Raman studies show that the toughening rubber particles dispersed in the cured vinylester resin are leached out during hygrothermal ageing. The weight gain during ageing suggests that this leaching process occurs concurrently with the absorption of water molecules. An increase in the degree of cross-linking is observed when bismaleimide/carbon fibre composite is aged at high temperature. This cross- linking tendency decreases with increasing depth within the carbon fibre bundle.

Sensing and using social context

We present online algorithms to extract social context: Social spheres are labeled locations of significance, represented as convex hulls extracted from GPS traces. Colocation is determined from Bluetooth and GPS to extract social rhythms, patterns in time, duration, place, and people corresponding to real-world activities. Social ties are formulated from proximity and shared spheres and rhythms. Quantitative evaluation is performed for 10+ million samples over 45 man-months. Applications are presented with assessment of perceived utility: Socio-Graph, a video and photo browser with filters for social metadata, and Jive, a blog browser that uses rhythms to discover similarity between entries automatically.

Dividing traffic sub-areas based on a parallel K-means algorithm

In order to alleviate the traffic congestion and reduce the complexity of traffic control and management, it is necessary to exploit traffic sub-areas division which should be effective in planing traffic. Some researchers applied the K-Means algorithm to divide traffic sub-areas on the taxi trajectories. However, the traditional K-Means algorithms faced difficulties in processing large-scale Global Position System(GPS) trajectories of taxicabs with the restrictions of memory, I/O, computing performance. This paper proposes a Parallel Traffic Sub-Areas Division(PTSD) method which consists of two stages, on the basis of the Parallel K-Means(PKM) algorithm. During the first stage, we develop a process to cluster traffic sub-areas based on the PKM algorithm. Then, the second stage, we identify boundary of traffic sub-areas on the base of cluster result. According to this method, we divide traffic sub-areas of Beijing on the real-word (GPS) trajectories of taxicabs. The experiment and discussion show that the method is effective in dividing traffic sub-areas.

Implementing the Flinders Model of self-management support with Aboriginal people who have diabetes : findings from a pilot study

A pilot program for Aboriginal people with diabetes on Eyre Peninsula, South Australia, aimed to test the acceptability and impact of using the Flinders model of self-management care planing to improve patient self-management. A community development approach was used to conduct a twelve-month demonstration project. Aboriginal health workers (AHWs) conducted patient-centred, self-management assessment and care planning. Impacts were measured by patient-completed diabetes self-management assessment tool, goal achievement, quality of life and clinical measures at baseline and 12 months. Impact and acceptability were also assessed by semi-structured interviews and focus groups of AHWs. Sixty Aboriginal people with type 2 diabetes stated their main problems as family and social dysfunction, access to services, nutrition and exercise. Problems improved by 12% and goals by 26%, while quality of life scores showed no significant change. Self-management scores improved in five of six domains. Mean HbA1c reduced from 8.74-8.09 and mean blood pressure was unchanged. AHWs found the process acceptable and appropriate for them and their patients. It was concluded that a diabetes self-management program provided by AHWs is acceptable, improves self-management and is seen to be useful by Aboriginal communities. Barriers include lack of preventative health services, social problems and time pressure on staff. Enablers include community concern regarding the prevalence and mortality associated with diabetes.