Exploiting SNOMED CT concepts and relationships for clinical information retrieval : Australian e-Health Research Centre and Queensland University of Technology at the TREC 2012 Medical Track

The Australian e-Health Research Centre and Queensland University of Technology recently participated in the TREC 2012 Medical Records Track. This paper reports on our methods, results and experience using an approach that exploits the concept and inter-concept relationships defined in the SNOMED CT medical ontology. Our concept-based approach is intended to overcome specific challenges in searching medical records, namely vocabulary mismatch and granularity mismatch. Queries and documents are transformed from their term-based originals into medical concepts as defined by the SNOMED CT ontology, this is done to tackle vocabulary mismatch. In addition, we make use of the SNOMED CT parent-child `is-a' relationships between concepts to weight documents that contained concept subsumed by the query concepts; this is done to tackle the problem of granularity mismatch. Finally, we experiment with other SNOMED CT relationships besides the is-a relationship to weight concepts related to query concepts. Results show our concept-based approach performed significantly above the median in all four performance metrics. Further improvements are achieved by the incorporation of weighting subsumed concepts, overall leading to improvement above the median of 28% infAP, 10% infNDCG, 12% R-prec and 7% Prec@10. The incorporation of other relations besides is-a demonstrated mixed results, more research is required to determined which SNOMED CT relationships are best employed when weighting related concepts.

Defining and describing what we do : doctrinal legal research

The practitioner lawyer of the past had little need to reflect on process. The doctrinal research methodology developed intuitively within the common law — a research method at the core of practice. There was no need to justify or classify it within a broader research framework. Modern academic lawyers are facing a different situation. At a time when competition for limited research funds is becoming more intense, and in which interdisciplinary work is highly valued and non-lawyers are involved in the assessment of grant applications, lawyer-applicants who engage in doctrinal research need to be able to explain their methodology more clearly. Doctrinal scholars need to be more open and articulate about their methods. These methods may be different in different contexts. This paper examines the doctrinal method used in legal research and its place in recent research dialogue. Some commentators are of the view that the doctrinal method is simply scholarship rather than a separate research methodology. Richard Posner even suggests that law is ‘not a field with a distinct methodology, but an amalgam of applied logic, rhetoric, economics and familiarity with a specialized vocabulary and a particular body of texts, practices, and institutions ...’.1 Therefore, academic lawyers are beginning to realise that the doctrinal research methodology needs clarification for those outside the legal profession and that a discussion about the standing and place of doctrinal research compared to other methodologies is required.

Learning personalized tag ontology from user tagging information

The cross-sections of the Social Web and the Semantic Web has put folksonomy in the spot light for its potential in overcoming knowledge acquisition bottleneck and providing insight for "wisdom of the crowds". Folksonomy which comes as the results of collaborative tagging activities has provided insight into user's understanding about Web resources which might be useful for searching and organizing purposes. However, collaborative tagging vocabulary poses some challenges since tags are freely chosen by users and may exhibit synonymy and polysemy problem. In order to overcome these challenges and boost the potential of folksonomy as emergence semantics we propose to consolidate the diverse vocabulary into a consolidated entities and concepts. We propose to extract a tag ontology by ontology learning process to represent the semantics of a tagging community. This paper presents a novel approach to learn the ontology based on the widely used lexical database WordNet. We present personalization strategies to disambiguate the semantics of tags by combining the opinion of WordNet lexicographers and users’ tagging behavior together. We provide empirical evaluations by using the semantic information contained in the ontology in a tag recommendation experiment. The results show that by using the semantic relationships on the ontology the accuracy of the tag recommender has been improved.

Personalization in tag ontology learning for recommendation making

Due to the explosive growth of the Web, the domain of Web personalization has gained great momentum both in the research and commercial areas. One of the most popular web personalization systems is recommender systems. In recommender systems choosing user information that can be used to profile users is very crucial for user profiling. In Web 2.0, one facility that can help users organize Web resources of their interest is user tagging systems. Exploring user tagging behavior provides a promising way for understanding users’ information needs since tags are given directly by users. However, free and relatively uncontrolled vocabulary makes the user self-defined tags lack of standardization and semantic ambiguity. Also, the relationships among tags need to be explored since there are rich relationships among tags which could provide valuable information for us to better understand users. In this paper, we propose a novel approach for learning tag ontology based on the widely used lexical database WordNet for capturing the semantics and the structural relationships of tags. We present personalization strategies to disambiguate the semantics of tags by combining the opinion of WordNet lexicographers and users’ tagging behavior together. To personalize further, clustering of users is performed to generate a more accurate ontology for a particular group of users. In order to evaluate the usefulness of the tag ontology, we use the tag ontology in a pilot tag recommendation experiment for improving the recommendation performance by exploiting the semantic information in the tag ontology. The initial result shows that the personalized information has improved the accuracy of the tag recommendation.

Rail incident investigator training: Is there support for a standardised Australasian training framework?

Background: Historically rail organisations have been operating in silos and devising their own training agendas. However with the harmonisation of the Australian workplace health and safety legislation and the appointment of a national rail safety regulator in 2013, rail incident investigator experts are exploring the possibility of developing a unified approach to investigator training. Objectives: The Australian CRC for Rail Innovation commissioned a training needs analysis to identify if common training needs existed between organisations and to assess support for the development of a national competency framework for rail incident investigations. Method: Fifty-two industry experts were consulted to explore the possibility of the development of a standardised training framework. These experts were sourced from within 19 Australasian organisations, comprising Rail Operators and Regulators in Queensland, New South Wales, Victoria, Western Australia, South Australia and New Zealand. Results: Although some competency requirements appear to be organisation specific, the vast majority of reported training requirements were generic across the Australasian rail operators and regulators. Industry experts consistently reported strong support for the development of a national training framework. Significance: The identification of both generic training requirements across organisations and strong support for standardised training indicates that the rail industry is receptive to the development of a structured training framework. The development of an Australasian learning framework could: increase efficiency in course development and reduce costs; establish recognised career pathways; and facilitate consistency with regards to investigator training.

Spatio temporal feature evaluation for action recognition

Spatio-Temporal interest points are the most popular feature representation in the field of action recognition. A variety of methods have been proposed to detect and describe local patches in video with several techniques reporting state of the art performance for action recognition. However, the reported results are obtained under different experimental settings with different datasets, making it difficult to compare the various approaches. As a result of this, we seek to comprehensively evaluate state of the art spatio- temporal features under a common evaluation framework with popular benchmark datasets (KTH, Weizmann) and more challenging datasets such as Hollywood2. The purpose of this work is to provide guidance for researchers, when selecting features for different applications with different environmental conditions. In this work we evaluate four popular descriptors (HOG, HOF, HOG/HOF, HOG3D) using a popular bag of visual features representation, and Support Vector Machines (SVM)for classification. Moreover, we provide an in-depth analysis of local feature descriptors and optimize the codebook sizes for different datasets with different descriptors. In this paper, we demonstrate that motion based features offer better performance than those that rely solely on spatial information, while features that combine both types of data are more consistent across a variety of conditions, but typically require a larger codebook for optimal performance.

Graph-based concept weighting for medical information retrieval

This paper presents a graph-based method to weight medical concepts in documents for the purposes of information retrieval. Medical concepts are extracted from free-text documents using a state-of-the-art technique that maps n-grams to concepts from the SNOMED CT medical ontology. In our graph-based concept representation, concepts are vertices in a graph built from a document, edges represent associations between concepts. This representation naturally captures dependencies between concepts, an important requirement for interpreting medical text, and a feature lacking in bag-of-words representations. We apply existing graph-based term weighting methods to weight medical concepts. Using concepts rather than terms addresses vocabulary mismatch as well as encapsulates terms belonging to a single medical entity into a single concept. In addition, we further extend previous graph-based approaches by injecting domain knowledge that estimates the importance of a concept within the global medical domain. Retrieval experiments on the TREC Medical Records collection show our method outperforms both term and concept baselines. More generally, this work provides a means of integrating background knowledge contained in medical ontologies into data-driven information retrieval approaches.

Of repetition and of song

Allyson Reynolds, for many years now, has been a keen student of nature. Nature’s forms have always been important to her work, but it would be wrong to see Reynolds as simply transcribing directly from the natural world. For this artist, nature and the natural world represents an elusive invitation and so Reynolds doesn’t so much paint from nature, although resemblance and imitation are clearly evident, as work to render visually and emotionally tangible this invitation. And for a viewer this is no small thing as the invitation is to become attuned and connect with nature as it exists in ourselves, that is to see as nature. Reynolds’ is inviting us as viewers to connect and participate with the nature of nature that is evident in ourselves through the act of perceiving. Reynolds’ work can do this because it is a work born of deep evocation. The sensibility from which it emerges or is transacted is poetic, intensely so, but more than that it works to make of viewing an act of poetry, through a celebration of seeing. Reynolds evocation is born of a deep feeling towards and contemplation of the natural world - it is as though the world lives in her and is now a seamless part of her creative vocabulary. In such personal work there is a profoundly felt and revealed sense of the intimate and like true intimacy it revels in both its dark and quiet, as well as its playful and light, aspects. What is special about Reynolds’ visual poetry is that it able to render this intimacy so accessible. It is an accessibility that is available for the viewer who is able to not only look but also surrender to that looking, accepting and working with the flow of thoughts and associations it occasions...

Personalized recommender systems integrating tags and item taxonomy

The social tags in Web 2.0 are becoming another important information source to profile users' interests and preferences to make personalized recommendations. To solve the problem of low information sharing caused by the free-style vocabulary of tags and the long tails of the distribution of tags and items, this paper proposes an approach to integrate the social tags given by users and the item taxonomy with standard vocabulary and hierarchical structure provided by experts to make personalized recommendations. The experimental results show that the proposed approach can effectively improve the information sharing and recommendation accuracy.

The effects of linguistic factors on student performance on economics multiple choice questions

This paper proposes a framework to analyse performance on multiple choice questions with the focus on linguistic factors. Item Response Theory (IRT) is deployed to estimate ability and question difficulty levels. A logistic regression model is used to detect Differential Item Functioning questions. Probit models testify relationships between performance and linguistic factors controlling the effects of question construction and students’ background. Empirical results have important implications. The lexical density of stems affects performance. The use of non-Economics specialised vocabulary has differing impacts on the performance of students with different language backgrounds. The IRT-based ability and difficulty help explain performance variations.

Basic cognitive experiences and definitions in the Longman Dictionary of Contemporary English

This study applies theories of cognitive linguistics to the compilation of English learners’ dictionaries. Specifically, it employs the concepts of basic level categories and image schemas, two basic cognitive experiences, to examine the ‘definition proper’ of English dictionaries for foreign learners. In the study, the definition proper refers to the constituent part of a reference work that provides an explanation of the meanings of a word, phrase or term. This rationalization mainly consists of defining vocabulary, sense division and arrangement, as well as the means of defining (i.e. paraphrase, true definition, functional definition, and pictorial illustration). The aim of the study is to suggest ways of aligning the consultation and learning of definitions with dictionary users’ cognitive experiences. For this purpose, an analysis of the definition proper of the fourth edition of the Longman Dictionary of Contemporary English (LDOCE4) from the perspective of basic cognitive experiences has been undertaken. The study found that, generally, the lexicographic practices of LDOCE4 are consistent with theories of cognitive linguistics. However, there exist shortcomings that result from disregarding basic cognitive experiences.

Which cultural values matter to business process management? Results from a global Delphi study

Purpose – Business process management (BPM) requires a holistic perspective that includes managing the culture of an organization to achieve objectives of efficient and effective business processes. Still, the specifics of a BPM-supportive organizational culture have not been examined so far. Thus, the purpose of this paper is to identify the characteristics of a cultural setting supportive of BPM objectives. Design/methodology/approach – The paper examines the constituent values of a BPM-supportive cultural setting through a global Delphi study with BPM experts from academia and practice and explore these values in a cultural value framework. Findings – The paper empirically identifies and defines four key cultural values supporting BPM, viz., customer orientation, excellence, responsibility, and teamwork. The paper discusses the relationships between these values and identifies a particular challenge in managing these seemingly competing values. Research limitations/implications – The identification and definition of these values represents a first step towards the operationalization (and empirical analysis) of what has been identified as the concept of BPM culture, i.e. a culture supportive of achieving BPM objectives. Practical implications – Identifying these cultural values provides the basis for developing an instrument that can measure how far an existing cultural context is supportive of BPM. This, in turn, is fundamental for identifying measures towards achieving a BPM culture as a necessary, yet not sufficient means to obtain BPM success. Originality/value – The paper examines which cultural values create an environment receptive for BPM and, thus, specifies the important theoretical construct BPM culture. In addition, the paper raises awareness for realizing these values in a BPM context.

Reproductive behavioural differences between wild-caught and pond-reared Penaeus monodon prawn broodstock

Ongoing problems exist with the commercial scale domestication of Penaeus monodon. One of the major issues, in terms of reproductive performance, is the low egg hatch rate of eggs from these captive bred prawns. The current study investigated the related issue of mating success. Time lapse video observations were conducted to compare the mating behaviour of pond-reared (domesticated) and wild-caught prawn P. monodon broodstock. Mating success of the pond-reared prawns was found to be low relative to wild-caught. It was determined that both male and female prawns contributed to this low mating rate suggesting both genders were impacted negatively by the domestication process. The causative factors for the low mating success are yet to be determined, however external physical abnormalities and lack of sexual maturity did not appear to play a role. The most notable behavioural difference between wild-caught and domesticated prawns was a reduced level of pursuit behaviour by domesticated males. This and other behavioural differences are discussed in relation to an increasing body of evidence that male prawns respond to sex pheromones produced by receptive females and that males detect these chemical signals in part, via their second antennal flagella. Accordingly we hypothesise that pond-reared (domesticated) females may have a reduced ability to produce or release sex pheromones and males, a reduced ability to detect them when compared to their wild-caught counterparts.

An ontology-based method for sparsity problem in tag recommendation

Tags or personal metadata for annotating web resources have been widely adopted in Web 2.0 sites. However, as tags are freely chosen by users, the vocabularies are diverse, ambiguous and sometimes only meaningful to individuals. Tag recommenders may assist users during tagging process. Its objective is to suggest relevant tags to use as well as to help consolidating vocabulary in the systems. In this paper we discuss our approach for providing personalized tag recommendation by making use of existing domain ontology generated from folksonomy. Specifically we evaluated the approach in sparse situation. The evaluation shows that the proposed ontology-based method has improved the accuracy of tag recommendation in this situation.

Child-related factors that influence teacher-child relationships using an Australian national sample

This study assessed the extent to which child-related factors had an impact on teacher-child relationships in Australian childcare settings. Analyses used data from Growing Up in Australia: The Longitudinal Study of Australian Children (LSAC). The sample consisted of 1577 two to three year old children (M= 33.9 months, SD=2.93; 51.5% male). Two separate hierarchical multiple regression analyses were conducted to examine the relation between teachers’ perceptions of their relationships with children and (a) gender, (b), indigenous status, (c), language background other than English, (d), socio-economic position, (e) special health care needs, (f) expression and receptive language concerns, (g) psychosocial competence and problems and (h) temperament factors (approach, persistence and reactivity). Results indicated that special health care needs, receptive language concerns and all three temperament scales (approach, persistence and reactivity) significantly predicted conflict in teacher-child relationships. Close relationships were predicted by being female, indigenous status, higher socio-economic position, not having a special health care need and no expressive language concerns.