Combining extension matrix and integer programming for optimal concept learning

This paper proposes two integer programming models and their GA-based solutions for optimal concept learning. The models are built to obtain the optimal concept description in the form of propositional logic formulas from examples based on completeness, consistency and simplicity. The simplicity of the propositional rules is selected as the objective function of the integer programming models, and the completeness and consistency of the concept are used as the constraints. Considering the real-world problems that certain level of noise is contained in data set, the constraints in model 11 are slacked by adding slack-variables. To solve the integer programming models, genetic algorithm is employed to search the global solution space. We call our approach IP-AE. Its effectiveness is verified by comparing the experimental results with other well- known concept learning algorithms: AQ15 and C4.5.

Near-surface distributions of soil water and water repellency under three effluent irrigation schemes in a blue gum (Eucalyptus globulus) plantation

Water repellent soils are difficult to irrigate and susceptible to preferential flow, which enhances the potential for accelerated leaching to groundwater of hazardous substances. Over 5 Mha of Australian soil is water repellent, while treated municipal sewage is increasingly used for irrigation. Only if a critical water content is exceeded will repellent soils become wettable. To avoid excessive loss of water from the root zone via preferential flow paths, irrigation schemes should therefore aim to keep the soil wet enough to maintain soil wettability. Our objective was to monitor the near-surface water content and water repellency in a blue gum (Eucalyptus globulus) plantation irrigated with treated sewage. The plantation's sandy soil surface was strongly water repellent when dry. For 4 months, three rows of 15 blue gum trees each received no irrigation, three other rows received 50% of the estimated potential water use minus rainfall, and three more rows received 100%. During this period, 162 soil samples were obtained in three sampling rounds, and their water content (% dry mass) and degree of water repellency determined. Both high and low irrigation effectively wetted up the soil and eliminated water repellency after 2 (high) or 4 (low) months. A single-peaked distribution of water contents was observed in the soil samples, but the water repellency distribution was dichotomous, with 44% extremely water-repellent and 36% wettable. This is consistent with a threshold water content at which a soil sample changes from water repellent to wettable, with spatial variability of this threshold creating a much wider transition zone at the field scale. We characterized this transition zone by expressing the fraction of wettable samples as a function of water content, and demonstrated a way to estimate from this the wettable portion of a field from a number of water content measurements. To keep the plantation soil wettable, the water content must be maintained at a level at which a significant downward flux is likely, with the associated enhanced leaching. At water contents with negligible downward flux, the field is water repellent, and leaching through preferential flow paths is likely. Careful management is needed to resolve these conflicting requirements.

An empirical study of encoding schemes and search strategies in discovering causal networks

Efficiently inducing precise causal models accurately reflecting given data sets is the ultimate goal of causal discovery. The algorithm proposed by Wallace et al. [10] has demonstrated its ability in discovering Linear Causal Models from data. To explore the ways to improve efficiency, this research examines three different encoding schemes and four searching strategies. The experimental results reveal that (1) specifying parents encoding method is the best among three encoding methods we examined; (2) In the discovery of linear causal models, local Hill climbing works very well compared to other more sophisticated methods, like Markov Chain Monte Carto (MCMC), Genetic Algorithm (GA) and Parallel MCMC searching.

An optimal strategy for extracting probabilistic rules by combining rough sets and genetic algorithm

This paper proposes an optimal strategy for extracting probabilistic rules from databases. Two inductive learning-based statistic measures and their rough set-based definitions: accuracy and coverage are introduced. The simplicity of a rule emphasized in this paper has previously been ignored in the discovery of probabilistic rules. To avoid the high computational complexity of rough-set approach, some rough-set terminologies rather than the approach itself are applied to represent the probabilistic rules. The genetic algorithm is exploited to find the optimal probabilistic rules that have the highest accuracy and coverage, and shortest length. Some heuristic genetic operators are also utilized in order to make the global searching and evolution of rules more efficiently. Experimental results have revealed that it run more efficiently and generate probabilistic classification rules of the same integrity when compared with traditional classification methods.

Supporting user evaluation of IT security certification schemes

IT Security Certification is an increasingly important qualification for information technology (IT) professionals seeking employment in IT security. Yet currently there is a lack of rigorously developed approaches to support the evaluation and selection by key stakeholders of the most appropriate IT security certification scheme from among hundreds of vendor-neutral and vendor-specific schemes. This paper develops a framework based on categories, characteristics and criteria to support user evaluation and selection of an (IT) Security Certification scheme that satisfies user priorities and requirements. The paper illustrates the use of the framework to support an experienced IT Professional’s evaluation. Theoretical and practical implications of the framework and trial evaluation are discussed.

An evaluation of the viability of photovoltaics in residential schemes managed by UK registered social landlords

Global demands on fossil fuels require the investigation of renewable and viable alternative energy supplies. The Intergovernmental Panel on Climate Change (IPCC) has concluded that current consumption of fossil fuels is untenable as atmospheric emissions of gases, in particular carbon dioxide (CO²), is having a significant and worsening effect on global climate change (IPCC 1992).

25% of UK CO² emissions are generated in the housing sector (UKCCP 2000). As major providers of UK social housing, Registered Social Landlords (RSLs), indirectly make a significant contribution to UK CO² emissions. In delivering UK Government policies, RSLs are required to meet national social and economic targets, as well as environmental targets. Clearly, social, environmental and economic issues combine in the arena of energy efficiency and social housing.

Potentially, the use of photovoltaics (PV) in social housing could assist the UK government in meeting targets in terms of affordable housing, providing "free" electricity to low income tenants, and with minimal environmental impact in urban areas. However, uptake of PV amongst RSLs in the UK has been minimal to date. This paper explores the factors that act as barriers to energy efficiency in this market.

Getting the complete picture : combining parental and child data to identify the barriers to social inclusion for children living in low socio-economic areas

Background: Childhood mental health problems are prevalent in Australian children (14–20%). Social exclusion is a risk factor for mental health problems, whereas being socially included can have protective effects. This study aims to identify the barriers to social inclusion for children aged 9–12 years living in low socio-economic status (SES) areas, using both child-report and parent-report interviews.

Methods: Australian-born English-speaking parents and children aged 9–12 years were sampled from a low SES area to participate in semi-structured interviews. Parents and children were asked questions around three prominent themes of social exclusion; exclusion from school, social activities and social networks.

Results: Many children experienced social exclusion at school, from social activities or within social networks. Overall, nine key barriers to social inclusion were identified through parent and child interviews, such as inability to attend school camps and participate in school activities, bullying and being left out, time and transport constraints, financial constraints and safety and traffic concerns. Parents and children often identified different barriers.

Discussion: There are several barriers to social inclusion for children living in low SES communities, many of which can be used to facilitate mental health promotion programmes. Given that parents and children may report different barriers, it is important to seek both perspectives.

Conclusion: This study strengthens the evidence base for the investments and action required to bring about the conditions for social inclusion for children living in low SES communities.

Extending the Workers' Compensation Act 1987 (N.S.W.) to include independent contractors and to allow more highly paid workers to insure fully their income, with particular reference to engineers

In mid-1987, the existing workers’ compensation system in New South Wales was replaced by a new Scheme, called ‘WorkCover’. While WorkCover solved a number of the financial problems that had plagued its predecessor, its enactment created other issues. Furthermore, WorkCover has failed to deal with a number of gaps in providing compensation for occupational injuries, most notably those suffered by independent contractors. By combining a study of aspects of industrial law and industrial relations, this thesis will examine some of those problems and gaps, in particular: (a) Should WorkCover be amended to enable independent contractors to come within its ambit? (b) Should there be additional insurance cover available (known as ‘top-up’ insurance) to insure those parts of workers’ wages presently left unprotected by WorkCover? (c) Should workers be permitted to take out another form of ‘top-up’ insurance to increase the quantum of death cover presently provided by the Scheme? (d) Should independent contractors who arc permitted to enter WorkCover also be permitted to obtain the extended cover set out in (b) and (c) above? Where appropriate, the thesis compares WorkCover to the workers’ compensation schemes in other Australian jurisdictions. It develops each of the matters referred to above by referring to the results of the writer’s survey of members of the Institution of Engineers (NSW Branch) which was conducted in May and June 1991.

A framework for assessing certification schemes for IT security professionals

This research develops a framework which allows the many IT security certifications to be compared by stakeholders, such as IT security professionals, employers, universities and governments. The framework employs a novel approach which allow users to tailor the comparison based on their own weightings, whilst taking advantage of standardised research.

Reliable algorithms and attack detection schemes for network systems

Network and Information security and reliability is still a key issue in information technology. This thesis develops two algorithms to improve the reliability and stability of content delivery systems, and proposes three attack detection schemes with high effectiveness and accuracy in detecting network attacks.

Authentication in public key encryption schemes

This thesis analyses authentication in public key encryption and makes contributions in two areas. Firstly, Compressed Nested PKI is proposed which improves the certificate validation process. Then an implicitly certified encryption scheme, Authenticated Public Key Encryption, which makes keys easier to manage, is introduced.