GYAN: A methodology for rule extraction from artificial neural networks

Artificial neural network (ANN) learning methods provide a robust and non-linear approach to approximating the target function for many classification, regression and clustering problems. ANNs have demonstrated good predictive performance in a wide variety of practical problems. However, there are strong arguments as to why ANNs are not sufficient for the general representation of knowledge. The arguments are the poor comprehensibility of the learned ANN, and the inability to represent explanation structures. The overall objective of this thesis is to address these issues by: (1) explanation of the decision process in ANNs in the form of symbolic rules (predicate rules with variables); and (2) provision of explanatory capability by mapping the general conceptual knowledge that is learned by the neural networks into a knowledge base to be used in a rule-based reasoning system. A multi-stage methodology GYAN is developed and evaluated for the task of extracting knowledge from the trained ANNs. The extracted knowledge is represented in the form of restricted first-order logic rules, and subsequently allows user interaction by interfacing with a knowledge based reasoner. The performance of GYAN is demonstrated using a number of real world and artificial data sets. The empirical results demonstrate that: (1) an equivalent symbolic interpretation is derived describing the overall behaviour of the ANN with high accuracy and fidelity, and (2) a concise explanation is given (in terms of rules, facts and predicates activated in a reasoning episode) as to why a particular instance is being classified into a certain category.

Compliant cryptologic protocols

Literally, the word compliance suggests conformity in fulfilling official requirements. The thesis presents the results of the analysis and design of a class of protocols called compliant cryptologic protocols (CCP). The thesis presents a notion for compliance in cryptosystems that is conducive as a cryptologic goal. CCP are employed in security systems used by at least two mutually mistrusting sets of entities. The individuals in the sets of entities only trust the design of the security system and any trusted third party the security system may include. Such a security system can be thought of as a broker between the mistrusting sets of entities. In order to provide confidence in operation for the mistrusting sets of entities, CCP must provide compliance verification mechanisms. These mechanisms are employed either by all the entities or a set of authorised entities in the system to verify the compliance of the behaviour of various participating entities with the rules of the system. It is often stated that confidentiality, integrity and authentication are the primary interests of cryptology. It is evident from the literature that authentication mechanisms employ confidentiality and integrity services to achieve their goal. Therefore, the fundamental services that any cryptographic algorithm may provide are confidentiality and integrity only. Since controlling the behaviour of the entities is not a feasible cryptologic goal,the verification of the confidentiality of any data is a futile cryptologic exercise. For example, there exists no cryptologic mechanism that would prevent an entity from willingly or unwillingly exposing its private key corresponding to a certified public key. The confidentiality of the data can only be assumed. Therefore, any verification in cryptologic protocols must take the form of integrity verification mechanisms. Thus, compliance verification must take the form of integrity verification in cryptologic protocols. A definition of compliance that is conducive as a cryptologic goal is presented as a guarantee on the confidentiality and integrity services. The definitions are employed to provide a classification mechanism for various message formats in a cryptologic protocol. The classification assists in the characterisation of protocols, which assists in providing a focus for the goals of the research. The resulting concrete goal of the research is the study of those protocols that employ message formats to provide restricted confidentiality and universal integrity services to selected data. The thesis proposes an informal technique to understand, analyse and synthesise the integrity goals of a protocol system. The thesis contains a study of key recovery,electronic cash, peer-review, electronic auction, and electronic voting protocols. All these protocols contain message format that provide restricted confidentiality and universal integrity services to selected data. The study of key recovery systems aims to achieve robust key recovery relying only on the certification procedure and without the need for tamper-resistant system modules. The result of this study is a new technique for the design of key recovery systems called hybrid key escrow. The thesis identifies a class of compliant cryptologic protocols called secure selection protocols (SSP). The uniqueness of this class of protocols is the similarity in the goals of the member protocols, namely peer-review, electronic auction and electronic voting. The problem statement describing the goals of these protocols contain a tuple,(I, D), where I usually refers to an identity of a participant and D usually refers to the data selected by the participant. SSP are interested in providing confidentiality service to the tuple for hiding the relationship between I and D, and integrity service to the tuple after its formation to prevent the modification of the tuple. The thesis provides a schema to solve the instances of SSP by employing the electronic cash technology. The thesis makes a distinction between electronic cash technology and electronic payment technology. It will treat electronic cash technology to be a certification mechanism that allows the participants to obtain a certificate on their public key, without revealing the certificate or the public key to the certifier. The thesis abstracts the certificate and the public key as the data structure called anonymous token. It proposes design schemes for the peer-review, e-auction and e-voting protocols by employing the schema with the anonymous token abstraction. The thesis concludes by providing a variety of problem statements for future research that would further enrich the literature.

Study of industrially relevant boundary layer and axisymmetric flows, including swirl and turbulence

Micropolar and RNG-based modelling of industrially relevant boundary layer and recirculating swirling flows is described. Both models contain a number of adjustable parameters and auxiliary conditions that must be either modelled or experimentally determined, and the effects of varying these on the resulting flow solutions is quantified. To these ends, the behaviour of the micropolar model for self-similar flow over a surface that is both stretching and transpiring is explored in depth. The simplified governing equations permit both analytic and numerical approaches to be adopted, and a number of closed form solutions (both exact and approximate) are obtained using perturbation and order of magnitude analyses. Results are compared with the corresponding Newtonian flow solution in order to highlight the differences between the micropolar and classical models, and significant new insights into the behaviour of the micropolar model are revealed for this flow. The behaviour of the RNG-bas based models for swirling flow with vortex breakdown zones is explored in depth via computational modelling of two experimental data sets and an idealised breakdown flow configuration. Meticulous modeling of upstream auxillary conditions is required to correctly assess the behavior of the models studied in this work. The novel concept of using the results to infer the role of turbulence in the onset and topology of the breakdown zone is employed.

The influence of habitat heterogeneity on patterns of connectivity among rabbit populations in southern Queensland

Patterns of connectivity among local populations influence the dynamics of regional systems, but most ecological models have concentrated on explaining the effect of connectivity on local population structure using dynamic processes covering short spatial and temporal scales. In this study, a model was developed in an extended spatial system to examine the hypothesis that long term connectivity levels among local populations are influenced by the spatial distribution of resources and other habitat factors. The habitat heterogeneity model was applied to local wild rabbit populations in the semi-arid Mitchell region of southern central Queensland (the Eastern system). Species' specific population parameters which were appropriate for the rabbit in this region were used. The model predicted a wide range of long term connectivity levels among sites, ranging from the extreme isolation of some sites to relatively high interaction probabilities for others. The validity of model assumptions was assessed by regressing model output against independent population genetic data, and explained over 80% of the variation in the highly structured genetic data set. Furthermore, the model was robust, explaining a significant proportion of the variation in the genetic data over a wide range of parameters. The performance of the habitat heterogeneity model was further assessed by simulating the widely reported recent range expansion of the wild rabbit into the Mitchell region from the adjacent, panmictic Western rabbit population system. The model explained well the independently determined genetic characteristics of the Eastern system at different hierarchic levels, from site specific differences (for example, fixation of a single allele in the population at one site), to differences between population systems (absence of an allele in the Eastern system which is present in all Western system sites). The model therefore explained the past and long term processes which have led to the formation and maintenance of the highly structured Eastern rabbit population system. Most animals exhibit sex biased dispersal which may influence long term connectivity levels among local populations, and thus the dynamics of regional systems. When appropriate sex specific dispersal characteristics were used, the habitat heterogeneity model predicted substantially different interaction patterns between female-only and combined male and female dispersal scenarios. In the latter case, model output was validated using data from a bi-parentally inherited genetic marker. Again, the model explained over 80% of the variation in the genetic data. The fact that such a large proportion of variability is explained in two genetic data sets provides very good evidence that habitat heterogeneity influences long term connectivity levels among local rabbit populations in the Mitchell region for both males and females. The habitat heterogeneity model thus provides a powerful approach for understanding the large scale processes that shape regional population systems in general. Therefore the model has the potential to be useful as a tool to aid in the management of those systems, whether it be for pest management or conservation purposes.

Discrete choice modelling of combined mode and departure time

Typical daily decision-making process of individuals regarding use of transport system involves mainly three types of decisions: mode choice, departure time choice and route choice. This paper focuses on the mode and departure time choice processes and studies different model specifications for a combined mode and departure time choice model. The paper compares different sets of explanatory variables as well as different model structures to capture the correlation among alternatives and taste variations among the commuters. The main hypothesis tested in this paper is that departure time alternatives are also correlated by the amount of delay. Correlation among different alternatives is confirmed by analyzing different nesting structures as well as error component formulations. Random coefficient logit models confirm the presence of the random taste heterogeneity across commuters. Mixed nested logit models are estimated to jointly account for the random taste heterogeneity and the correlation among different alternatives. Results indicate that accounting for the random taste heterogeneity as well as inter-alternative correlation improves the model performance.

Conifer defence against insects: microarray gene expression profiling of Sitka spruce (Picea sitchensis) induced by mechanical wounding or feeding by spruce budworms (Choristoneura occidentalis) or white pine weevils (Pissodes strobi) reveals large

Conifers are resistant to attack from a large number of potential herbivores or pathogens. Previous molecular and biochemical characterization of selected conifer defence systems support a model of multigenic, constitutive and induced defences that act on invading insects via physical, chemical, biochemical or ecological (multitrophic) mechanisms. However, the genomic foundation of the complex defence and resistance mechanisms of conifers is largely unknown. As part of a genomics strategy to characterize inducible defences and possible resistance mechanisms of conifers against insect herbivory, we developed a cDNA microarray building upon a new spruce (Picea spp.) expressed sequence tag resource. This first-generation spruce cDNA microarray contains 9720 cDNA elements representing c. 5500 unique genes. We used this array to monitor gene expression in Sitka spruce (Picea sitchensis) bark in response to herbivory by white pine weevils (Pissodes strobi, Curculionidae) or wounding, and in young shoot tips in response to western spruce budworm (Choristoneura occidentalis, Lepidopterae) feeding. Weevils are stem-boring insects that feed on phloem, while budworms are foliage feeding larvae that consume needles and young shoot tips. Both insect species and wounding treatment caused substantial changes of the host plant transcriptome detected in each case by differential gene expression of several thousand array elements at 1 or 2 d after the onset of treatment. Overall, there was considerable overlap among differentially expressed gene sets from these three stress treatments. Functional classification of the induced transcripts revealed genes with roles in general plant defence, octadecanoid and ethylene signalling, transport, secondary metabolism, and transcriptional regulation. Several genes involved in primary metabolic processes such as photosynthesis were down-regulated upon insect feeding or wounding, fitting with the concept of dynamic resource allocation in plant defence. Refined expression analysis using gene-specific primers and real-time PCR for selected transcripts was in agreement with microarray results for most genes tested. This study provides the first large-scale survey of insect-induced defence transcripts in a gymnosperm and provides a platform for functional investigation of plant-insect interactions in spruce. Induction of spruce genes of octadecanoid and ethylene signalling, terpenoid biosynthesis, and phenolic secondary metabolism are discussed in more detail.

Bioreactor Systems for Tissue Engineering

The hydrodynamic environment “created” by bioreactors for the culture of a tissue engineered construct (TEC) is known to influence cell migration, proliferation and extra cellular matrix production. However, tissue engineers have looked at bioreactors as black boxes within which TECs are cultured mainly by trial and error, as the complex relationship between the hydrodynamic environment and tissue properties remains elusive, yet is critical to the production of clinically useful tissues. It is well known in the chemical and biotechnology field that a more detailed description of fluid mechanics and nutrient transport within process equipment can be achieved via the use of computational fluid dynamics (CFD) technology. Hence, the coupling of experimental methods and computational simulations forms a synergistic relationship that can potentially yield greater and yet, more cohesive data sets for bioreactor studies. This review aims at discussing the rationale of using CFD in bioreactor studies related to tissue engineering, as fluid flow processes and phenomena have direct implications on cellular response such as migration and/or proliferation. We conclude that CFD should be seen by tissue engineers as an invaluable tool allowing us to analyze and visualize the impact of fluidic forces and stresses on cells and TECs.

Potential soil carbon sequestration in overgrazed grassland ecosystems

Excessive grazing pressure is detrimental to plant productivity and may lead to declines in soil organic matter. Soil organic matter is an important source of plant nutrients and can enhance soil aggregation, limit soil erosion, and can also increase cation exchange and water holding capacities, and is, therefore, a key regulator of grassland ecosystem processes. Changes in grassland management which reverse the process of declining productivity can potentially lead to increased soil C. Thus, rehabilitation of areas degraded by overgrazing can potentially sequester atmospheric C. We compiled data from the literature to evaluate the influence of grazing intensity on soil C. Based on data contained within these studies, we ascertained a positive linear relationship between potential C sequestration and mean annual precipitation which we extrapolated to estimate global C sequestration potential with rehabilitation of overgrazed grassland. The GLASOD and IGBP DISCover data sets were integrated to generate a map of overgrazed grassland area for each of four severity classes on each continent. Our regression model predicted losses of soil C with decreased grazing intensity in drier areas (precipitation less than 333 mm yr(-1)), but substantial sequestration in wetter areas. Most (93%) C sequestration potential occurred in areas with MAP less than 1800 mm. Universal rehabilitation of overgrazed grasslands can sequester approximately 45 Tg C yr(-1), most of which can be achieved simply by cessation of overgrazing and implementation of moderate grazing intensity. Institutional level investments by governments may be required to sequester additional C.

Preliminary modelling outcomes : calculation of holding costs in greenfield residential development

Many of the costs associated with greenfield residential development are apparent and tangible. For example, regulatory fees, government taxes, acquisition costs, selling fees, commissions and others are all relatively easily identified since they represent actual costs incurred at a given point in time. However, identification of holding costs are not always immediately evident since by contrast they characteristically lack visibility. One reason for this is that, for the most part, they are typically assessed over time in an ever-changing environment. In addition, wide variations exist in development pipeline components: they are typically represented from anywhere between a two and over sixteen years time period - even if located within the same geographical region. Determination of the starting and end points, with regards holding cost computation, can also prove problematic. Furthermore, the choice between application of prevailing inflation, or interest rates, or a combination of both over time, adds further complexity. Although research is emerging in these areas, a review of the literature reveals attempts to identify holding cost components are limited. Their quantification (in terms of relative weight or proportionate cost to a development project) is even less apparent; in fact, the computation and methodology behind the calculation of holding costs varies widely and in some instances completely ignored. In addition, it may be demonstrated that ambiguities exists in terms of the inclusion of various elements of holding costs and assessment of their relative contribution. Yet their impact on housing affordability is widely acknowledged to be profound, with their quantification potentially maximising the opportunities for delivering affordable housing. This paper seeks to build on earlier investigations into those elements related to holding costs, providing theoretical modelling of the size of their impact - specifically on the end user. At this point the research is reliant upon quantitative data sets, however additional qualitative analysis (not included here) will be relevant to account for certain variations between expectations and actual outcomes achieved by developers. Although this research stops short of cross-referencing with a regional or international comparison study, an improved understanding of the relationship between holding costs, regulatory charges, and housing affordability results.

Curbing paper wastage using flavoured feedback

In November 2009 the researcher embarked on a project aimed at reducing the amount of paper used by Queensland University of Technology (QUT) staff in their daily workplace activities. The key goal was to communicate to staff that excessive printing has a tangible and negative effect on their workplace and local environment. The research objective was to better understand what motivates staff towards more ecologically sustainable printing practises, whilst meeting their job’s demands. The current study is built on previous research that found that one interface does not address the needs of all users when creating persuasive Human Computer Interaction (HCI) interventions targeting resource consumption. In response, the current study created and trialled software that communicates individual paper consumption in precise metrics. Based on preliminary research data different metric sets have been defined to address the different motivations and beliefs of user archetypes using descriptive and injunctive normative information.

A preliminary examination of the effects of changes in motorcycle licensing in Queensland

Several sets of changes have been made to motorcycle licensing in Queensland since 2007, with the aim of improving the safety of novice riders. These include a requirement that a motorcycle learner licence applicant must have held a provisional or open car licence for 12 months, and imposing a 3 year limit for learner licence renewal. Additionally, a requirement to hold an RE (250 cc limited) class licence for a period of 12 months prior to progressing to an R class licence was introduced for Q-RIDE. This paper presents analyses of licensing transaction data that examine the effects of the licensing changes on the duration that the learner licence was held, factors affecting this duration and the extent to which the demographic characteristics of learner licence holders changed. The likely safety implications of the observed changes are discussed.

Pockets of creativity in business processes

Creative processes, for instance, the development of visual effects or computer games, increasingly become part of the agenda of information systems researchers and practitioners. Such processes get their managerial challenges from the fact that they comprise both well-structured, transactional parts and creative parts. The latter can often not be precisely specified in terms of control flow, required resources, and outcome. The processes’ high uncertainty sets boundaries for the application of traditional business process management concepts, such as process automation, process modeling, process performance measurement, and risk management. Organizations must thus exercise caution when it comes to managing creative processes and supporting these with information technology. This, in turn, requires a profound understanding of the concept of creativity in business processes. In response to this, the present paper introduces a framework for conceptualizing creativity within business processes. The conceptual framework describes three types of uncertainty and constraints as well as the interrelationships among these. The study is grounded in the findings from three case studies that were conducted in the film and visual effects industry. Moreover, we provide initial evidence for the framework’s validity beyond this narrow focus. The framework is intended to serve as a sensitizing device that can guide further information systems research on creativity-related phenomena.

Statistical and Econometric Methods for Transportation Data Analysis, 2nd edition

Now in its second edition, this book describes tools that are commonly used in transportation data analysis. The first part of the text provides statistical fundamentals while the second part presents continuous dependent variable models. With a focus on count and discrete dependent variable models, the third part features new chapters on mixed logit models, logistic regression, and ordered probability models. The last section provides additional coverage of Bayesian statistical modeling, including Bayesian inference and Markov chain Monte Carlo methods. Data sets are available online to use with the modeling techniques discussed.

Household type and structure, time-use pattern, and trip-chaining behavior

In order to examine time allocation patterns within household-level trip-chaining, simultaneous doubly-censored Tobit models are applied to model time-use behavior within the context of household activity participation. Using the entire sample and a sub-sample of worker households from Tucson's Household Travel Survey, two sets of models are developed to better understand the phenomena of trip-chaining behavior among five types of households: single non-worker households, single worker households, couple non-worker households, couple one-worker households, and couple two-worker households. Durations of out-of-home subsistence, maintenance, and discretionary activities within trip chains are examined. Factors found to be associated with trip-chaining behavior include intra-household interactions with the household types and their structure and household head attributes.

Reliability evaluations of railway power supplies by fault-tree analysis

High reliability of railway power systems is one of the essential criteria to ensure quality and cost-effectiveness of railway services. Evaluation of reliability at system level is essential for not only scheduling maintenance activities, but also identifying reliability-critical components. Various methods to compute reliability on individual components or regularly structured systems have been developed and proven to be effective. However, they are not adequate for evaluating complicated systems with numerous interconnected components, such as railway power systems, and locating the reliability critical components. Fault tree analysis (FTA) integrates the reliability of individual components into the overall system reliability through quantitative evaluation and identifies the critical components by minimum cut sets and sensitivity analysis. The paper presents the reliability evaluation of railway power systems by FTA and investigates the impact of maintenance activities on overall reliability. The applicability of the proposed methods is illustrated by case studies in AC railways.