Text mining in negative relevance feedback

It is a big challenge to clearly identify the boundary between positive and negative streams. Several attempts have used negative feedback to solve this challenge; however, there are two issues for using negative relevance feedback to improve the effectiveness of information filtering. The first one is how to select constructive negative samples in order to reduce the space of negative documents. The second issue is how to decide noisy extracted features that should be updated based on the selected negative samples. This paper proposes a pattern mining based approach to select some offenders from the negative documents, where an offender can be used to reduce the side effects of noisy features. It also classifies extracted features (i.e., terms) into three categories: positive specific terms, general terms, and negative specific terms. In this way, multiple revising strategies can be used to update extracted features. An iterative learning algorithm is also proposed to implement this approach on RCV1, and substantial experiments show that the proposed approach achieves encouraging performance.

Is the unigram relevance model term independent? Classifying term dependencies in query expansion

This paper develops a framework for classifying term dependencies in query expansion with respect to the role terms play in structural linguistic associations. The framework is used to classify and compare the query expansion terms produced by the unigram and positional relevance models. As the unigram relevance model does not explicitly model term dependencies in its estimation process it is often thought to ignore dependencies that exist between words in natural language. The framework presented in this paper is underpinned by two types of linguistic association, namely syntagmatic and paradigmatic associations. It was found that syntagmatic associations were a more prevalent form of linguistic association used in query expansion. Paradoxically, it was the unigram model that exhibited this association more than the positional relevance model. This surprising finding has two potential implications for information retrieval models: (1) if linguistic associations underpin query expansion, then a probabilistic term dependence assumption based on position is inadequate for capturing them; (2) the unigram relevance model captures more term dependency information than its underlying theoretical model suggests, so its normative position as a baseline that ignores term dependencies should perhaps be reviewed.

Document clustering evaluation : Divergence from a random baseline

Divergence from a random baseline is a technique for the evaluation of document clustering. It ensures cluster quality measures are performing work that prevents ineffective clusterings from giving high scores to clusterings that provide no useful result. These concepts are defined and analysed using intrinsic and extrinsic approaches to the evaluation of document cluster quality. This includes the classical clusters to categories approach and a novel approach that uses ad hoc information retrieval. The divergence from a random baseline approach is able to differentiate ineffective clusterings encountered in the INEX XML Mining track. It also appears to perform a normalisation similar to the Normalised Mutual Information (NMI) measure but it can be applied to any measure of cluster quality. When it is applied to the intrinsic measure of distortion as measured by RMSE, subtraction from a random baseline provides a clear optimum that is not apparent otherwise. This approach can be applied to any clustering evaluation. This paper describes its use in the context of document clustering evaluation.

Reading images : the book club in the art gallery

This paper seeks to document and understand one instance of community-university engagement: that of an on-going book club organised in conjunction with public art exhibitions. The curator of the Queensland University of Technology (QUT) Art Museum invited the authors, three postgraduate research students in the faculty of Creative Writing and Literary Studies at QUT, to facilitate an informal book club. The purpose of the book club was to generate discussion, through engagement with fiction, around the themes and ideas explored in the Art Museum’s exhibitions. For example, during the William Robinson exhibition, which presented evocative images of the environment around Brisbane, Queensland, the book club explored texts that symbolically represented aspects of the Australian landscape in a variety of modes and guises. This paper emerges as a result of the authors’ observations during, and reflections on, their experiences facilitating the book club. It responds to the research question, how can we create a best practice model to engage readers through open-ended, reciprocal discussion of fiction, while at the same time encouraging interactions in the gallery space? To provide an overview of reading practices in book clubs, we rely on Jenny Hartley’s seminal text on the subject, The Reading Groups Book (2002). Although the book club was open to all members of the community, the participants were generally women. Elizabeth Long, in Book Clubs: Woman and the Uses of Reading in the Everyday (2003), offers a comprehensive account of women’s interactions as they engage in a reading community. Long (2003, 2) observes that an image of the solitary reader governs our understanding of reading. Long challenges this notion, arguing that reading is profoundly social (ibid), and, as women read and talk in book clubs, ‘they are supporting each other in a collective working-out of their relationship to a particular historical movement and the particular social conditions that characterise it’ (Long 2003, 22). Despite the book club’s capacity to act as a forum for analytical discussion, DeNel Rehberg Sedo (2010, 2) argues that there are barriers to interaction in such a space, including that members require a level of cultural capital and literacy before they feel comfortable to participate. How then can we seek to make book clubs more inclusive, and encourage readers to discuss and question outside of their comfort zone? How can we support interactions with texts and images? In this paper, we draw on pragmatic and self-reflective practice methods to document and evaluate the development of the book club model designed to facilitate engagement. We discuss how we selected texts, negotiating the dual needs of relevance to the exhibition and engagement with, and appeal to, the community. We reflect on developing questions and material prior to the book club to encourage interaction, and describe how we developed a flexible approach to question-asking and facilitating discussion. We conclude by reflecting on the outcomes of and improvements to the model.

Integration of medical images to the teaching of systematic pathology : an evaluation of relevance

A sound knowledge of pathological disease processes is required for professional practice within health professions. The project described in this paper reviewed the resources currently available for the delivery of systematic pathology tutorials. Additional complementary resources were developed and the inclusion of these additional learning resources in practical tutorial sessions was evaluated for their impact on student learning. Student evaluation of the learning resources was undertaken across one semester with two different cohorts of health profession students using questionnaires and focus group discussion. Both cohorts reported an enhancement to their understanding of pathological disease processes through the use of the additional resources. Results indicate student perception of the value of the resources correlates with staff perception and is independent of prior experiences.

Examination of socio-demographics and job satisfaction in Australian registered nurses

Aims and Objectives To examine relationships between socio-demographic factors and job satisfaction and to identify if these factors predict job satisfaction levels in an Australian registered nurses. Background Reports indicate that in Australia there are 30,000 qualified nurses no longer working in the healthcare and that current nursing shortages vary as a result of certain socio-demographic variables including type of nurse, geographic, location, sector, service and organisation. Furthermore it has been revealed that there is not only a real shortage but also a pseudo-shortage (i.e. either there are not enough nurses are available, or not enough are willing to work under existing workplace conditions). International studies have found significant relationships exist between some socio-demographic factors and job satisfaction in registered nurses however there is limited information available on relationships between socio-demographic factors and job satisfaction in nurses in the Australian context. Design A cross sectional survey was undertaken of Australian registered nurses. Methods Two thousand Australian registered nurses who were members of an industrial and professional organisation were sent the questionnaire in 2008. They were stratified and randomised according to gender. Six hundred and thirty-nine registered nurses responded. Descriptive analyses, correlation analyses, one- way ANOVA tests, simple linear regression and multivariable analyses were conducted to examine further if any relationships existed between the variables. Results The majority of respondents showed positive job satisfaction scores. An ANOVA found significant positive relationships existed between job satisfaction, specialty area, health sector and Australian states. Multivariable analyses found relationships existed between specialty area, health sector, and job satisfaction. Conclusions The variables specialty area and health sector were found to be significantly associated with job satisfaction. The different specialty areas and health sectors in relation to job satisfaction should be investigated further. Clinical Relevance The study results have provided new knowledge for policy makers, organisational and nursing leaders of the socio-demographic variables that may affect job satisfaction in registered nurses in the Australian context.

Tourist use of mopeds in Queensland

Tourist use of mopeds in Queensland is encouraged by licensing regulations permitting moped riding for car licence holders, who may lack prior knowledge or experience of moped or motorcycle use. Using official crash and registration data, this research examines moped use by tourists, identified as crash-involved riders holding an interstate or overseas licence. Tourists were more likely to be younger, female, in single vehicle crashes, and deemed at fault than Queensland licence holders. Potential safety issues include poor riding skills, inexperience, inattention and lack of protective clothing. Moped rental companies play an important role in managing client crash and injury risks. These risks could also be reduced through introduction of more stringent licensing requirements, though this may be detrimental to moped rental companies as well as to tourist mobility and enjoyment. The discussion considers the relevance of adventure tourism perspectives and theory to the use of mopeds by tourists.

Types of foot problems seen by Australian podiatrists

Background: Understanding frequency of foot problems can assist health care planners with resource deployment to new and emerging services such as paediatric podiatry and focus future research on the most salient foot conditions. Methods: A review of 2187 patient consultations during a three month period was conducted. Patient medical and podiatric history was coded using industry standards. All patients were recruited for convenience from a metropolitan university podiatry clinic. Results: 392 new patients were identified with mean age 40.6 years old (range 1–95), with 65% being female. Arthritic diseases, asthma, hypertension and allergies were the most common medical conditions reported. The frequency of new consultations in younger people (n = 102; 27%) exceeded those of the elderly (n = 75; 20%). Conversely, the elderly were nearly three times more prevalent in this cohort (n = 910; 43%) compared to younger people (n = 332; 16%). Conclusion: This study illustrates the diverse nature of pathology seen by podiatrists. Knowledge that skin lesions are highly prevalent is of relevance to health departments, given the aging nature of most populations. Moreover there appears to be a growing trend in the number of young people who present for care, however government funded access to these services are limited.

The cancer stem-cell hypothesis : its emerging role in lung cancer biology and its relevance for future therapy

The cancer stem-cell (CSC) hypothesis suggests that there is a small subset of cancer cells that are responsible for tumor initiation and growth, possessing properties such as indefinite self-renewal, slow replication, intrinsic resistance to chemotherapy and radiotherapy, and an ability to give rise to differentiated progeny. Through the use of xenotransplantation assays, putative CSCs have been identified in many cancers, often identified by markers usually expressed in normal stem cells. This is also the case in lung cancer, and the accumulated data on side population cells, CD133, CD166, CD44 and ALDH1 are beginning to clarify the true phenotype of the lung cancer stem cell. Furthermore, it is now clear that many of the pathways of normal stem cells, which guide cellular proliferation, differentiation, and apoptosis are also prominent in CSCs; the Hedgehog (Hh), Notch, and Wnt signaling pathways being notable examples. The CSC hypothesis suggests that there is a small reservoir of cells within the tumor, which are resistant to many standard therapies, and can give rise to new tumors in the form of metastases or relapses after apparent tumor regression. Therapeutic interventions that target CSC pathways are still in their infancy and clinical data of their efficacy remain limited. However Smoothened inhibitors, gamma-secretase inhibitors, anti-DLL4 antagonists, Wnt antagonists, and CBP/β-catenin inhibitors have all shown promising anticancer effects in early studies. The evidence to support the emerging picture of a lung cancer CSC phenotype and the development of novel therapeutic strategies to target CSCs are described in this review.

The continuing relevance of the library as a third place for users and non-users of IT : the case of Canada Bay

The purpose of this paper is to consider how libraries support the development of community networks both physically and digitally. To do this, a case-study methodology was employed, including a combination of data about the library and qualitative interviews with library users considering their experience of the library. This paper proposes that libraries act as ‘third places’ spatially connecting people; libraries also build links with online media and play a critical role in inclusively connecting non-technology users with the information on the Internet and digital technology more generally. The paper establishes the value of libraries in the digital age and recommends that libraries actively seek ways to develop links between non-technology users and activity on the Internet. It addresses the need to reach these types of non-technology users in different ways. Further, it suggests that libraries utilise their positioning as third places to create broader community networks, to support local communities beyond existing users and beyond the library precinct.

Retrieving information from microblog using pattern mining and relevance feedback

Retrieving information from Twitter is always challenging due to its large volume, inconsistent writing and noise. Most existing information retrieval (IR) and text mining methods focus on term-based approach, but suffers from the problems of terms variation such as polysemy and synonymy. This problem deteriorates when such methods are applied on Twitter due to the length limit. Over the years, people have held the hypothesis that pattern-based methods should perform better than term-based methods as it provides more context, but limited studies have been conducted to support such hypothesis especially in Twitter. This paper presents an innovative framework to address the issue of performing IR in microblog. The proposed framework discover patterns in tweets as higher level feature to assign weight for low-level features (i.e. terms) based on their distributions in higher level features. We present the experiment results based on TREC11 microblog dataset and shows that our proposed approach significantly outperforms term-based methods Okapi BM25, TF-IDF and pattern based methods, using precision, recall and F measures.

Overview of the INEX 2011 relevance feedback track

The INEX 2011 Relevance Feedback track offered a refined approach to the evaluation of Focused Relevance Feedback algorithms through simulated exhaustive user feedback. Run in largely identical fashion to the Relevance Feedback track in INEX 2010[2], we simulated a user-in-the loop by re-using the assessments of ad-hoc retrieval obtained from real users who assess focused ad-hoc retrieval submissions. We present the evaluation methodology, its implementation, and experimental results obtained for four submissions from two participating organisations. As the task and evaluation methods did not change between INEX 2010 and now, explanations of these details from the INEX 2010 version of the track have been repeated verbatim where appropriate.

Animating and sustaining niche social networks

Within the communicative space online Social Network Sites (SNS) afford, Niche Social Networks Sites (NSNS) have emerged around particular geographic, demographic or topic-based communities to provide what broader SNS do not: specified and targeted content for an engaged and interested community. Drawing on a research project developed at the Queensland University of Technology in conjunction with the Australian Smart Services Cooperative Research Centre that produced an NSNS based around Adventure Travel, this paper outlines the main drivers for community creation and sustainability within NSNS. The paper asks what factors motivate users to join and stay with these sites and what, if any, common patterns can be noted in their formation. It also outlines the main barriers to online participation and content creation in NSNS, and the similarities and differences in SNS and NSNS business models. Having built a community of 100 registered members, the staywild.com.au project was a living laboratory, enabling us to document the steps taken in producing a NSNS and cultivating and retaining active contributors. The paper incorporates observational analysis of user-generated content (UGC) and user profile submissions, statistical analysis of site usage, and findings from a survey of our membership pool in noting areas of success and of failure. In drawing on our project in this way we provide a template for future iterations of NSNS initiation and development across various other social settings: not only niche communities, but also the media and advertising with which they engage and interact. Positioned within the context of online user participation and UGC research, our paper concludes with a discussion of the ways in which the tools afforded by NSNS extend earlier understandings of online ‘communities of interest’. It also outlines the relevance of our research to larger questions about the diversity of the social media ecology.

Adopting relevance feature to learn personalized ontologies

Relevance feature and ontology are two core components to learn personalized ontologies for concept-based retrievals. However, how to associate user native information with common knowledge is an urgent issue. This paper proposes a sound solution by matching relevance feature mined from local instances with concepts existing in a global knowledge base. The matched concepts and their relations are used to learn personalized ontologies. The proposed method is evaluated elaborately by comparing it against three benchmark models. The evaluation demonstrates the matching is successful by achieving remarkable improvements in information filtering measurements.

Semantic labelling for document feature patterns using ontological subjects

Finding and labelling semantic features patterns of documents in a large, spatial corpus is a challenging problem. Text documents have characteristics that make semantic labelling difficult; the rapidly increasing volume of online documents makes a bottleneck in finding meaningful textual patterns. Aiming to deal with these issues, we propose an unsupervised documnent labelling approach based on semantic content and feature patterns. A world ontology with extensive topic coverage is exploited to supply controlled, structured subjects for labelling. An algorithm is also introduced to reduce dimensionality based on the study of ontological structure. The proposed approach was promisingly evaluated by compared with typical machine learning methods including SVMs, Rocchio, and kNN.