Andy Laws

Autonomic system design based on the integrated use of SSM and VSM

This paper describes an approach to autonomic systems development that involves the integrated use of two general systems design techniques: Soft Systems Methodology (SSM) and the Viable System Model (VSM). The paper summarizes the relevant aspects of each technique and describes how they can be used together to create design models of an autonomic system and its environment. The discussion is illustrated through a consideration of the development of a generic computing system to help manage the technology used in an organisation.

Adjustable deliberation of self-managing systems

In this paper a cybernetics-based viable system architectural model is introduced, which provides a blueprint for high-assurance systems and a meta-control model necessary for the adjustable deliberation and autonomy of self-managing systems. The logical formalism is provided by the Situation Calculus and underpinned by an Enhanced Belief-Desires-Intentions (EBDI) framework to facilitate the representation and reasoning of scaleable autonomic computing systems.

From wetware to software: a cybernetic perspective of self-adaptive software

The development and application of software engineering practices over the last thirty years have undoubtedly resulted in the production of significantly improved software. However, the majority of modem software systems remain intrinsically fragile and highly vulnerable to environmental change and require continuing and problematic manual adaptation. In this paper and given the problems inherent in manual software adaptation, the authors argue that imbuing the software system with the ability to self-adapt offers a potentially profitable route forward. For support of this claim, the authors draw on the emerging discipline of self-adaptive software, which seeks to devolve some of the responsibility for maintenance activity to the software itself. Realizing such auto-adaptive capability proves to be a challenging problem. The authors contend that many of the themes, problems and goals currently identified in the field of self-adaptive software bear a striking resemblance to problems that have long formed the basis of enquiry in the well-established field of cybernetics. Classical cybernetics, drawing on mathematical models of the adaptive processes of biological organisms, seeks to identify the general principles of control and communication required for organisms to survive in a changing environment. Consequently, cybernetics appears to offer the potential to apply naturally developed adaptation strategies to software artifacts. Therefore, after discussing these theoretical foundations, this paper reports their practical application by presenting the initial findings from the development of an experimental, agent based, adaptive In-Vehicle Telematics System (IVTS) for use by the Emergency Services.

Genetically modified software: realizing viable autonomic agency

Inspired by the autonomic aspects of the human central nervous system, the vision of autonomic computing arrived with a fully-formed wish list of characteristics that such systems should exhibit, essentially those self-referential aspects required for effective self-management. Here, the authors contend that the biologically-inspired managerial cybernetics of Beers Viable System Model (VSM) provides significant conceptual guidance for the development of a general architecture for the operation and management of such complex, evolving, adaptive systems. Consequently, the VSM has been used as the basis of a theoretically-supported reference model that provides the blueprint for an extensible intelligent agent architecture. Of course, normal use of the VSM relies heavily on human agency to realize the adaptive capabilities required by the model. Therefore, artificially replicating such activities represents a significant challenge, however the authors show that some progress can be made using algorithmic hot swapping and in particular Hollands Genetic Algorithms (GAs) to generate, in specific circumstances, a repertoire of tailored responses to environmental change. The authors then speculate on the use of the associated Learning Classifier Systems (LCS) approach to allow the system to develop an adaptive environmental model of appropriate, optimized responses.

Website design: Technical, social and medical issues for self-reporting by elderly patients

There is growing interest in the use of the Internet for interacting with patients, both in terms of healthcare information provision and information gathering. In this article, we examine the issues in designing healthcare websites for elderly users. In particular, this article uses a year-long case study of the development of a web-based system for self-reporting of symptoms and quality of life with a view to examine the issues relating to website design for elderly users. The issues identified included the technical, social and medical aspects of website design for elderly users. The web-based system developed was based on the European Quality of Life 5-Dimensions health-status questionnaire, a commonly used tool for patient self-reporting of quality of life, and the more specific coronary revascularisation outcome questionnaire. Currently, self-reporting is generally administered in the form of paper-based questionnaires to be completed in the outpatient clinic or at home. There are a variety of issues relating to elderly users, which imply that websites for elderly patients may involve different design considerations to other types of websites.

Towards viable computer systems: a set theory interpretation of ecological dependence within Beer's self-organizing viable system model

Presented is research articulating a novel technology progressing resource management within self-organizing systems. Examining both Cybernetic and Autonomic Computing techniques we evolve a set-theory oriented, atomically-derived, emergent model that reflects an algorithmic decomposition of Beers recursive, multi-agent Viable System Model, pertinent by its composition of multiple and independent entities, sharing one or more objectives. Integrated management promotes each sub-system as a whole within a closed ecological meta-boundary. The relationships between sub-systems is demonstrated via syntax subscripts, while the relationship linking recursive levels is recognized via superscripts. The resultant design grammar endorses autonomy versus governance, exploiting cybernetic, biological and mathematical metaphors, crucially seeking inherent learning and control through system-environment interplay. Focusing on interactions and inter-relationships, the self-organizing environments exhibit evolution of systemic elements, conserving yet managing resources provided by each entity. Research ultimately aspires augm entation of the Autonomic Computing state of the art into the original field of Viable Computing Systems.

Machine learning approaches to predict learning outcomes in Massive open online courses

With the rapid advancements in technology, Massive Open Online Courses (MOOCs) have become the most popular form of online educational delivery, largely due to the removal of geographical and financial barriers for participants. A large number of learners globally enrol in such courses. Despite the flexible accessibility, results indicate that the completion rate is quite low. Educational Data Mining and Learning Analytics are emerging fields of research that aim to enhance the delivery of education through the application of various statistical and machine learning approaches. An extensive literature survey indicates that no significant research is available within the area of MOOC data analysis, in particular considering the behavioural patterns of users. In this paper, therefore, two sets of features, based on learner behavioural patterns, were compared in terms of their suitability for predicting the course outcome of learners participating in MOOCs. Our Exploratory Data Analysis demonstrates that there is strong correlation between click stream actions and successful learner outcomes. Various Machine Learning algorithms have been applied to enhance the accuracy of classifier models. Simulation results from our investigation have shown that Random Forest achieved viable performance for our prediction problem, obtaining the highest performance of the models tested. Conversely, Linear Discriminant Analysis achieved the lowest relative performance, though represented only a marginal reduction in performance relative to the Random Forest.

Issues in online patient self-reporting of health status

Patient self-reporting of symptoms and quality of life following surgical interventions is generally delivered in the form of paper-based questionnaires to be completed in the outpatient clinic or at home. A commonly used tool for patient self-reporting of quality of life is the EQ5D health status questionnaire which, while limited in scope, has general applicability across a range of health interventions. In this article we examine the issues relating to online patient self-reporting using this questionnaire and the wider implications for the online reporting of health status.

An open-bounded cybernetic case study of viable computing systems: Applying directive correlation to an algorithmic hot-swapping scenario

This paper presents an open-bounded case study of our Viable Computer System (VCS), the design grammar model innovating a hybrid VCS architectural representation of Beers cybernetic Viable System Model (VSM). When applied to a previous genetically-modified system scenario, System One represents a metaphor for homeostasis. The set-theoretical framework defines research specifics, i.e. systems and their environments via algorithmic hot-swapping. Further functions and a set of disturbances are introduced, supplying a potential repertoire of tailored responses to open environmental change. Fundamental to promoting emergence, thus viability is Sommerhoffs concept of directive correlation and Ashbys notion of goal-directedness, i.e. the ability to achieve a goal-state under variations in the environment. Example identities exhibit potential for context-free portability including sets of values of environmental and behavioral variables and a set of outcomes allowing the system to develop an adaptive environmental model of fit responses illustrating temporal and autonomic properties of the VCS concept.

Towards the Differentiation of Initial and Final Retention in Massive Open Online Courses

Following an accelerating pace of technological change, Massive Open Online Courses (MOOCs) have emerged as a popular educational delivery platform, leveraging ubiquitous connectivity and computing power to overcome longstanding geographical and financial barriers to education. Consequently, the demographic reach of education delivery is extended towards a global online audience, facilitating learning and development for a continually expanding portion of the world population. However, an extensive literature review indicates that the low completion rate is a major issue related to MOOCs. This is considered to be a lack of person to person interaction between instructors and learners on such courses and, the ability of tutors to monitor learners is impaired, often leading to learner withdrawals. To address this problem, learner drop out patterns across five courses offered by Harvard and MIT universities are investigated in this paper. Learning Analytics is applied to address key factors behind participant dropout events through the comparison of attrition during the first and last weeks of each course. The results show that the attrition of participants during the first week of the course is higher than during the last week, low percentages of learners’ attrition are found prior to course closing dates. This could indicate that assessment fees may not represent a significant reason for learners’ withdrawal. We introduce supervised machine learning algorithms for the analysis of learner retention and attrition within a MOOC platform. Results show that machine learning represents a viable direction for the predictive analysis of MOOCs outcomes, with the highest performances yielded by Boosted Tree classification for initial attrition and Neural Network based classification for final attrition.