Classification of cancer-related death certificates using machine learning

Background Cancer monitoring and prevention relies on the critical aspect of timely notification of cancer cases. However, the abstraction and classification of cancer from the free-text of pathology reports and other relevant documents, such as death certificates, exist as complex and time-consuming activities. Aims In this paper, approaches for the automatic detection of notifiable cancer cases as the cause of death from free-text death certificates supplied to Cancer Registries are investigated. Method A number of machine learning classifiers were studied. Features were extracted using natural language techniques and the Medtex toolkit. The numerous features encompassed stemmed words, bi-grams, and concepts from the SNOMED CT medical terminology. The baseline consisted of a keyword spotter using keywords extracted from the long description of ICD-10 cancer related codes. Results Death certificates with notifiable cancer listed as the cause of death can be effectively identified with the methods studied in this paper. A Support Vector Machine (SVM) classifier achieved best performance with an overall F-measure of 0.9866 when evaluated on a set of 5,000 free-text death certificates using the token stem feature set. The SNOMED CT concept plus token stem feature set reached the lowest variance (0.0032) and false negative rate (0.0297) while achieving an F-measure of 0.9864. The SVM classifier accounts for the first 18 of the top 40 evaluated runs, and entails the most robust classifier with a variance of 0.001141, half the variance of the other classifiers. Conclusion The selection of features significantly produced the most influences on the performance of the classifiers, although the type of classifier employed also affects performance. In contrast, the feature weighting schema created a negligible effect on performance. Specifically, it is found that stemmed tokens with or without SNOMED CT concepts create the most effective feature when combined with an SVM classifier.

Application of machine learning technique in wind turbine fault diagnosis

In this study, a machine learning technique called anomaly detection is employed for wind turbine bearing fault detection. Basically, the anomaly detection algorithm is used to recognize the presence of unusual and potentially faulty data in a dataset, which contains two phases: a training phase and a testing phase. Two bearing datasets were used to validate the proposed technique, fault-seeded bearing from a test rig located at Case Western Reserve University to validate the accuracy of the anomaly detection method, and a test to failure data of bearings from the NSF I/UCR Center for Intelligent Maintenance Systems (IMS). The latter data set was used to compare anomaly detection with SVM, a previously well-known applied method, in rapidly finding the incipient faults.

Developing agency in the creative career : a design-based framework for work integrated learning

Research shows that approximately half of creative practitioners operate as embedded creatives by securing gainful employment within organisations located in the field beyond their core discipline. This foregrounds the significance of having the skills necessary to successfully cross the disciplinary boundaries in order to negotiate a professional role. The multiple implications of such reframing for emerging creative practitioners who navigate uncertain professional boundaries include developing a skill of identifying and successfully targeting the shifting professional and industry coordinates while remaining responsive to changes. A further implication involves creative practitioners engaging in a continuous cycle of re-negotiation of their professional identity making the management of multiple professional selves - along with creating and recreating a meaningful frame of references such as the language around their emerging practice - a necessary skill. This chapter presents a case study of a set of Work Integrated Learning subjects designed to develop in creative industries practitioners the skills to manage their emerging professional identities in response to the shifts in the professional world.

Standards that we can 'own' : Threshold Learning Outcomes for Journalism, Public Relations, and Media & Communication Studies

This paper outlines the progress by the JoMeC (Journalism, Media & Communication) Network in developing TLO (Threshold Learning Outcome) statements for Bachelor-level university programs in the disciplines of Journalism, Public Relations and Media & Communications Studies. The paper presents the finalised TLO statement for Journalism, and outlines moves to engage discipline-based groups to further develop preliminary TLOs for Public Relations and Media & Communication Studies. The JoMeC Network was formed in 2011, in response to requirements that from 2014 all degrees and qualifications at Australian universities would be able to demonstrate that they comply with the threshold learning standards set by the Australian Qualifications Framework (AQF). The AQF’s threshold standards define the minimum types and levels of knowledge, skills and capabilities that a student must demonstrate in order to graduate. The Tertiary Education Quality and Standards Agency (TEQSA) will use the AQF’s threshold standards as a key tool in recording and assessing the performance of higher educational institutions, and determining whether they should be registered as Australian Higher Education Providers under the Higher Education Standards Framework. The Office of Learning & Teaching (OLT) places the onus on discipline communities to collaborate in order to develop and ‘own’ the threshold learning standards that can be considered the minimum learning outcomes of university-level programs in that field. With the support of an OLT Grant, the JoMeC Network’s prime goal has been to develop three sets of discipline-specific TLOs – one each for the Journalism, Public Relations, and Media & Communications Studies disciplines. This paper describes the processes of research, consultation, drafting and ongoing revision of the TLO for Journalism. It outlines the processes that the JoMeC Network has taken in developing a preliminary TLO draft to initiate discussion of Public Relations and Media & Communication Studies. The JoMeC Network plans to hand management of further development of these TLOs to scholars within the discipline who will engage with academics and other stakeholders to develop statements that the respective disciplines can embrace and ‘own’.

Factors influencing robustness and effectiveness of conditional random fields in active learning frameworks

Active learning approaches reduce the annotation cost required by traditional supervised approaches to reach the same effectiveness by actively selecting informative instances during the learning phase. However, effectiveness and robustness of the learnt models are influenced by a number of factors. In this paper we investigate the factors that affect the effectiveness, more specifically in terms of stability and robustness, of active learning models built using conditional random fields (CRFs) for information extraction applications. Stability, defined as a small variation of performance when small variation of the training data or a small variation of the parameters occur, is a major issue for machine learning models, but even more so in the active learning framework which aims to minimise the amount of training data required. The factors we investigate are a) the choice of incremental vs. standard active learning, b) the feature set used as a representation of the text (i.e., morphological features, syntactic features, or semantic features) and c) Gaussian prior variance as one of the important CRFs parameters. Our empirical findings show that incremental learning and the Gaussian prior variance lead to more stable and robust models across iterations. Our study also demonstrates that orthographical, morphological and contextual features as a group of basic features play an important role in learning effective models across all iterations.

Learning analytics beyond the LMS : the Connected Learning Analytics toolkit

We present a Connected Learning Analytics (CLA) toolkit, which enables data to be extracted from social media and imported into a Learning Record Store (LRS), as defined by the new xAPI standard. Core to the toolkit is the notion of learner access to their own data. A number of implementational issues are discussed, and an ontology of xAPI verb/object/activity statements as they might be unified across 7 different social media and online environments is introduced. After considering some of the analytics that learners might be interested in discovering about their own processes (the delivery of which is prioritised for the toolkit) we propose a set of learning activities that could be easily implemented, and their data tracked by anyone using the toolkit and a LRS.

Implementing the self-management threshold learning outcome for law : some intentional design strategies from the current curriculum toolbox

There is a growing awareness of the high levels of psychological distress being experienced by law students and the practising profession in Australia. In this context, a Threshold Learning Outcome (TLO) on self-management has been included in the six TLOs recently articulated as minimum learning outcomes for all Australian graduates of the Bachelor of Laws degree (LLB). The TLOs were developed during 2010 as part of the Australian Learning and Teaching Council’s (ALTC’s) project funded by the Australian Government to articulate ‘Learning and Teaching Academic Standards’. The TLOs are the result of a comprehensive national consultation process led by the ALTC’s Discipline Scholars: Law, Professors Sally Kift and Mark Israel.1 The TLOs have been endorsed by the Council of Australian Law Deans (CALD) and have received broad support from members of the judiciary and practising profession, representative bodies of the legal profession, law students and recent graduates, Legal Services Commissioners and the Law Admissions Consultative Committee. At the time of writing, TLOs for the Juris Doctor (JD) are also being developed, utilising the TLOs articulated for the LLB as their starting point but restating the JD requirements as the higher order outcomes expected of graduates of a ‘Masters Degree (Extended)’, this being the award level designation for the JD now set out in the new Australian Qualifications Framework.2 As Australian law schools begin embedding the learning, teaching and assessment of the TLOs in their curricula, and seek to assure graduates’ achievement of them, guidance on the implementation of the self-management TLO is salient and timely.

From music student to industry professional: An entrepreneurial learning design

This design-based research project addresses the gap between formal music education curricula and the knowledge and skills necessary to enter the professional music industry. It analyses the work of a teacher/researcher who invited her high school students to start their own business venture, Youth Music Industries (YMI). YMI also functioned as a learning environment informed by the theoretical concepts of communities of practice and social capital. The students staged cycles of events of various scales over a three-year period, as platforms for young artists to engage and develop new, young audiences across Queensland, Australia. The study found that students developed an entrepreneurial mindset through acquisition of specific skills and knowledge. Their learning was captured and distilled into a set of design principles, a pedagogical approach transferrable across the creative industries more broadly.

Do-Rank: DCG optimization for learning-to-rank in tag-based item recommendation systems

Discounted Cumulative Gain (DCG) is a well-known ranking evaluation measure for models built with multiple relevance graded data. By handling tagging data used in recommendation systems as an ordinal relevance set of {negative,null,positive}, we propose to build a DCG based recommendation model. We present an efficient and novel learning-to-rank method by optimizing DCG for a recommendation model using the tagging data interpretation scheme. Evaluating the proposed method on real-world datasets, we demonstrate that the method is scalable and outperforms the benchmarking methods by generating a quality top-N item recommendation list.

Designing hubs for connected learning: social, spatial and technological insights from Coworking, Hackerspaces and Meetup groups

Connected learning, as a design approach, does not restrict learning to a dedicated learning space (school, university, etc.), but considers it to be an aggregation of individual experiences made through intrinsically motivated, active participation in and across various socio-cultural, every-day life environments. Urban places for meeting, interacting and connected learning with people from diverse backgrounds, cultures and areas of expertise are highly significant in the knowledge economy of our 21st century. However, little is yet known about best practices to design and curate such hubs that attract and support interest-driven and socially embedded learning experiences. The research study presented in this paper investigates design aspects that contribute to successful place-based spaces for connected learning. The paper reports findings from observations as well as interviews with users and managers of three different types of local, community-led learning environments, i.e., coworking spaces, hackerspaces, and meetup groups across Australia. The findings reveal social, spatial and technological interventions that these spaces apply to nourish a culture of connected learning, sharing and peer interaction. The discussion suggests a set of design implications for designers, managers and decision makers that have an interest in nourishing a connected learning culture among their user community.

Teaching scenario analysis – An action learning pedagogy

In this paper we provide an introduction to our teaching of scenario analysis. Scenario analysis offers an excellent instructional vehicle for investigating ‘wicked problems’; issues that are complex and ambiguous and require trans-disciplinary inquiry. We outline the pedagogical underpinning based on action learning and provide a critical approach from the intuitive logics school of scenario analysis. We use this in our programme in which student groups engage in semi-structured, but divergent and inclusive analysis of a selected focal issue. They then develop a set of scenario storylines that outline the limits of possibility and plausibility for a selected time-horizon year. The scenarios are portrayed not as narratives, but as vehicles for exploration of the causes and outcomes of the interplay between forces in the contextual environment that drive the unfolding future in the context of the focal issue. In this way, we provide internally-generated challenges to both individual pre-conceptions and group-level thinking.

Negotiating Chineseness through learning Chinese as a Heritage Language in Australia: The role of habitus

While education researchers have drawn on the work of a wide diversity of theorists over the years, much contemporary theory building in these areas has revolved around the work of Pierre Bourdieu. Theory as Method in Research develops the capacity of students, researchers and teachers to successfully put Bourdieu’s ideas to work in their own research and prepare them effectively for conducting Masters and Doctoral scholarships. Structured around four core themes, this book provides a range of research case studies exploring educational identities, educational inequalities, school leadership and management, and research in teacher education. Issues as diverse as Chinese language learning and identity, school leadership in Australia and the school experience of Afro-Trinidadian boys, are covered, intertwined with a set of innovative approaches to theory application in education research. This collection brings together, in one comprehensive volume, a set of education researchers who place Pierre Bourdieu’s key concepts such as habitus, capital and field at the centre of their research methodologies. Full of insight and innovation, the book is an essential read for practitioners, student teachers, researchers and academics who want to harness the potential of Bourdieu’s core concepts in their own work, thereby helping to bridge the gap between theory and method in education research.

Rethinking Culture and Diversity in Early Childhood Preservice Teacher Education: Looking at Teaching and Learning Through an Interdisciplinary Lens

Early Childhood Education (ECE) has a long history of building foundations for children to achieve their full potential, enabling parents to participate in the economy while children are cared for, addressing poverty and disadvantage, and building individual, community and societal resources. In so doing, ECE has developed a set of cultural practices and ways of knowing that shape the field and the people who work within it. ECE, consequently, is frequently described as unique and special (Moss, 2006; Penn, 2011). This works to define and distinguish the field while, simultaneously, insulating it from other contexts, professions, and ideas. Recognising this dualism illuminates some of the risks and challenges of operating in an insular and isolated fashion. In the 21st century, there are new challenges for children, families and societies to which ECE must respond if it is to continue to be relevant. One major issue is how ECE contributes to transition towards more sustainable ways of living. Addressing this contemporary social problem is one from which Early Childhood teacher education has been largely absent (Davis & Elliott, 2014), despite the well recognised but often ignored role of education in contributing to sustainability. Because of its complexity, sustainability is sometimes referred to as a ‘wicked problem’ (Rittel & Webber, 1973; Australian Public Service Commission, 2007) requiring alternatives to ‘business as usual’ problem solving approaches. In this chapter, we propose that addressing such problems alongside disciplines other than Education enables the Early Childhood profession to have its eyes opened to new ways of thinking about our work, potentially liberating us from the limitations of our “unique” and idiosyncratic professional cultures. In our chapter, we focus on understandings of culture and diversity, looking to broaden these by exploring the different ‘cultures’ of the specialist fields of ECE and Design (in this project, we worked with students studying Architecture, Industrial Design, Landscape Architecture and Interior Design). We define culture not as it is typically represented, i.e. in relation to ideas and customs of particular ethnic and language groups, but to the ideas and practices of people working in different disciplines and professions. We assert that different specialisms have their own ‘cultural’ practices. Further, we propose that this kind of theoretical work helps us to reconsider ways in which ECE might be reframed and broadened to meet new challenges such as sustainability and as yet unknown future challenges and possibilities. We explore these matters by turning to preservice Early Childhood teacher education (in Australia) as a context in which traditional views of culture and diversity might be reconstructed. We are looking to push our specialist knowledge boundaries and to extend both preservice teachers and academics beyond their comfort zones by engaging in innovative interdisciplinary learning and teaching. We describe a case study of preservice Early Childhood teachers and designers working in collaborative teams, intersecting with a ‘real-world’ business partner. The joint learning task was the design of an early learning centre based on sustainable design principles and in which early Education for Sustainability (EfS) would be embedded Data were collected via focus group and individual interviews with students in ECE and Design. Our findings suggest that interdisciplinary teaching and learning holds considerable potential in dismantling taken-for-granted cultural practices, such that professional roles and identities might be reimagined and reconfigured. We conclude the chapter with provocations challenging the ways in which culture and diversity in the field of ECE might be reconsidered within teacher education.

Affective learning design: A principled approach to emotion in learning

This PhD project set out to explore the role of emotion during learning in sport, focusing on how actions, emotions and cognitions interact under the influence of constraints. Key outcomes include the development of the theoretical concept - Affective Learning Design, and a new tool for assessing the intensity of emotions during learning - the Sport Learning and Emotions Questionnaire. The findings presented in this thesis provide both theoretical and practical implications discussing why emotion should be considered in the design of learning environments in sport.

The evolution of entrepreneurial learning

- Purpose The purpose of this paper is to present an evolutionary perspective on entrepreneurial learning, whilst also accounting for fundamental ecological processes, by focusing on the development of key recurring, knowledge components within nascent and growing small businesses. - Design/methodology/approach The paper relates key developments within the organizational evolution literature to research on entrepreneurial learning, with arguments presented in favor of adopting a multi‐level co‐evolutionary perspective that captures and explains hidden ecological process, such as niche‐construction. - Findings It is argued in the paper that such a multi‐level focus on key recurring knowledge components can shed new light on the process of entrepreneurial learning and lead to the cross‐fertilization of ideas across different domains of study, by offering researchers the opportunity to use the framework of variation‐selection‐retention to develop a multi‐level representation of organizational and entrepreneurial learning. - Originality/value Entrepreneurial learning viewed in this way, as a multi‐level struggle for survival amongst competing knowledge components, can provide entrepreneurs with a set of evolutionary heuristics as they re‐interpret their understanding of the evolution of their business.