Medical semantic similarity with a neural language model

Advances in neural network language models have demonstrated that these models can effectively learn representations of words meaning. In this paper, we explore a variation of neural language models that can learn on concepts taken from structured ontologies and extracted from free-text, rather than directly from terms in free-text. This model is employed for the task of measuring semantic similarity between medical concepts, a task that is central to a number of techniques in medical informatics and information retrieval. The model is built with two medical corpora (journal abstracts and patient records) and empirically validated on two ground-truth datasets of human-judged concept pairs assessed by medical professionals. Empirically, our approach correlates closely with expert human assessors ($\approx$ 0.9) and outperforms a number of state-of-the-art benchmarks for medical semantic similarity. The demonstrated superiority of this model for providing an effective semantic similarity measure is promising in that this may translate into effectiveness gains for techniques in medical information retrieval and medical informatics (e.g., query expansion and literature-based discovery).

Approximate nearest-neighbour search with inverted signature slice lists

In this paper we present an original approach for finding approximate nearest neighbours in collections of locality-sensitive hashes. The paper demonstrates that this approach makes high-performance nearest-neighbour searching feasible on Web-scale collections and commodity hardware with minimal degradation in search quality.

Parallel streaming signature EM-tree: A clustering algorithm for web scale applications

The proliferation of the web presents an unsolved problem of automatically analyzing billions of pages of natural language. We introduce a scalable algorithm that clusters hundreds of millions of web pages into hundreds of thousands of clusters. It does this on a single mid-range machine using efficient algorithms and compressed document representations. It is applied to two web-scale crawls covering tens of terabytes. ClueWeb09 and ClueWeb12 contain 500 and 733 million web pages and were clustered into 500,000 to 700,000 clusters. To the best of our knowledge, such fine grained clustering has not been previously demonstrated. Previous approaches clustered a sample that limits the maximum number of discoverable clusters. The proposed EM-tree algorithm uses the entire collection in clustering and produces several orders of magnitude more clusters than the existing algorithms. Fine grained clustering is necessary for meaningful clustering in massive collections where the number of distinct topics grows linearly with collection size. These fine-grained clusters show an improved cluster quality when assessed with two novel evaluations using ad hoc search relevance judgments and spam classifications for external validation. These evaluations solve the problem of assessing the quality of clusters where categorical labeling is unavailable and unfeasible.

"Unruly women and semiotic spaces": Jay Verney's a mortality tale

Written at a time of significant changes for women and for women writers in Queensland, Australia, Jay Verney's A Mortality Tale (1994), privileges women's experiences of place and seeks to redeem the feminine from its entrapment in masculine stories/discourses of self and place. This chapter draws on Julia Kristeva's conceptualisation of the semiotic as a way of reading Verney's witty and reflective re-articulation of phallocentric orderings of spatiality and gender.

Exploring the Potential to Educate “Good Citizens” through the Australian Civics and Citizenship Curriculum.

This chapter examines the ways in which notions of ‘a good citizen’ and ‘civic virtue’ have been conceptualized in the new Civics and Citizenship Curriculum for students in Years 3 – 10 in Australia. It argues that whilst Civics and Citizenship Education (CCE) has, over time and in various ways, been recognized as a significant aspect of Australian education, only recently has attention been given to the relational and multidimensional conceptions of citizenship. Considerations of ‘morality’, ‘a good citizen’ and ‘civic virtue’ offer possibilities to engage with multidimensional notions of citizenship, which acknowledge that citizenship perspectives can be affected by personal, social, spatial and temporary situations (Cogan & Derricott, 2000). In the current statement on national goals for schooling in Australia, which informed the development of CCE, the Melbourne Declaration (MCEETYA, 2008) called for young Australians to be educated to “act with moral and ethical integrity” and be “committed to national values of democracy, equity and justice, and participate in Australia’s civic life” (MCEETYA, 2008, pp. 8–9). The chapter claims that this maximal emphasis (McLaughlin, 1992), based on active, values based and interpretive approaches to democratic citizenship which encourage debate and participation in civil society, was evident in the new Civics and Citizenship Curriculum. However, it contends that the recommendations of the recent Review of the Australian Curriculum: Final report (Australian Government, 2014a & b), will now limit CCE’s potential to deliver the sort of active and informed citizenship heralded by the Melbourne Declaration. This is because the Review advocates for a content-focused minimal (McLaughlin, 1992) emphasis on civic knowledge, with diminished attention to citizenship participation and processes. In doing so, the Review foregrounds conceptions of the ‘good citizen’ in more limited terms of responsibility, obligations and compliance with the status quo.

From News Blogs to News on Twitter: Gatewatching and Collaborative News Curation

This chapter, explores the role of the second tier of independent news blogs as it developed in the years following the Seattle WTO protests in 1999, and outlines the practice of gatewatching as a key element of news bloggers’ activities. We critique perceptions of the news blogosphere as an echo chamber or filter bubble whose discussions about current events are detached from journalistic coverage, and demonstrate instead the close interconnections between independent news bloggers and professional journalists in the wider media ecology. Finally, we sketch the gradual transition and broadening of gatewatching practices in the news blogosphere towards the collaborative curation of news sharing in contemporary social media spaces, and outline the further research questions which emerge from such transformations of the flows of news and discussion.

Live and analogue : Theatre as a practice of social inclusion

This monograph argues for a repositioning of theatre as an antidote to the negative effects of the reduction of live social interaction among the current generation of young people. It argues for a repositioned valuing of the role that live performing arts can play in the development of social cohesion and well-being with young people. Highlighted therefore is an emphasis on developing ‘communitas’ or ‘collective joy’ (Turner, 2012) for young audiences. Using the lens of social acupuncture (O’Donnell, 2006) the monograph explores how two Australian live theatre or performance events – The Walking Neighbourhood and 지하 Underground – foreground inclusionary audience experiences. It concludes with a challenge to drama teachers, theatre makers, arts researchers, arts programmers and performance venue managers regarding how they might view their young audiences as more than just a marketing demographic and income stream, by providing new platforms for young people’s social cohesion and well-being.

Text segmentation in unconstrained hand-drawings in whiteboard photos

In this paper we present a robust method to detect handwritten text from unconstrained drawings on normal whiteboards. Unlike printed text on documents, free form handwritten text has no pattern in terms of size, orientation and font and it is often mixed with other drawings such as lines and shapes. Unlike handwritings on paper, handwritings on a normal whiteboard cannot be scanned so the detection has to be based on photos. Our work traces straight edges on photos of the whiteboard and builds graph representation of connected components. We use geometric properties such as edge density, graph density, aspect ratio and neighborhood similarity to differentiate handwritten text from other drawings. The experiment results show that our method achieves satisfactory precision and recall. Furthermore, the method is robust and efficient enough to be deployed in a mobile device. This is an important enabler of business applications that support whiteboard-centric visual meetings in enterprise scenarios. © 2012 IEEE.

No Right to Live? Malaysia’s Islam and Implications for Its Sexual Minority Storytellers

This chapter discusses the fast emerging challenges for Malay and Muslim sexual minority storytellers in the face of an aggressive state-sponsored Islamisation of a constitutionally secular Malaysia. I examine the case of Azwan Ismail, a gay Malay and Muslim Malaysian who took part in the local ‘It Gets Better’ project, and who suffered an onslaught of hostile comments from fellow Malay Muslims. Azwan’s experience makes one question how a message of discouraging suicidal tendencies among sexual minority youths can be so vehemently misperceived. Azwan’s existential challenges – stemming from the tension between his own constructions of self and those of others – (re)present a unique challenge in the long struggle for human rights. In my examination of the arising contradictions, I highlight the challenges for Azwan’s existential self – one who is deemed morally bankrupt by hostile audiences. The purist Sunni Islam agenda in a constitutionally secular Malaysia not only rejects the human rights of the sexual minorities in Malaysia but has also influenced, and is often a leading hostile voice in both regional and international blocs. This self-righteous, supremacist and authoritarian Islam discourages discourse and attacks all differing opinions. This resulting disabling environment for vulnerable, minority communities and their human rights manifests in State-endorsed discrimination, compulsory counselling, forced rehabilitation and criminalisation. It places the rights of the sexual minorities to live within such a society in doubt. In discussing the arising issues, I draw upon literature that investigates the way in which personal stories have traditionally been used to advance human rights. Included too, is the significance and implications of the work by social psychologists in explaining the loss of credibility of personal stories. I then advance an analytical framework that will allow storytelling as a very individual form of witnessing to reclaim and regain its ‘truth to power’.

Public Policy for the Creative Industries

The cultural and creative industries are closely intertwined with government. This chapter reviews key economic rationales for public policy interventions for the arts, cultural and creative industries. Market failure justifications depend on the status of arts and culture as non-rival public goods, as ‘merit goods’, or the need to moderate the effects of up-front investment costs or monopoly, and the inherent uncertainty of creative production. ‘Systems failure’ too is a regular rationale for policy intervention. Using the United Kingdom as an example, the chapter shows how emphasis on these rationales has shifted over the last three decades, first in the context of industrial policies for traditional aims such as exports and job growth, which have been joined in recent years by the need for investment in intangibles, knowledge exchange, and spillover effects in the wider economy.

Quality-aware review selection based on product feature taxonomy

As of today, user-generated information such as online reviews has become increasingly significant for customers in decision making process. Meanwhile, as the volume of online reviews proliferates, there is an insistent demand to help the users tackle the information overload problem. In order to extract useful information from overwhelming reviews, considerable work has been proposed such as review summarization and review selection. Particularly, to avoid the redundant information, researchers attempt to select a small set of reviews to represent the entire review corpus by preserving its statistical properties (e.g., opinion distribution). However, one significant drawback of the existing works is that they only measure the utility of the extracted reviews as a whole without considering the quality of each individual review. As a result, the set of chosen reviews may consist of low-quality ones even its statistical property is close to that of the original review corpus, which is not preferred by the users. In this paper, we proposed a review selection method which takes review quality into consideration during the selection process. Specifically, we examine the relationships between product features based upon a domain ontology to capture the review characteristics based on which to select reviews that have good quality and preserve the opinion distribution as well. Our experimental results based on real world review datasets demonstrate that our proposed approach is feasible and able to improve the performance of the review selection effectively.

The iconic and the critical

Theodor Adorno was opposed to the cinema because he felt it was too close to reality, and ipso facto an extension of ideological Capital, as he wrote in 1944 in Dialectic of Enlightenment. What troubled Adorno was the iconic nature of cinema – the semiotic category invented by C. S. Peirce where the signifier (sign) does not merely signify, in the arbitrary capacity attested by Saussure, but mimics the formal-visual qualities of its referent. Iconicity finds its perfect example in the film’s ingenuous surface illusion of an unmediated reality – its genealogy (the iconic), since classical antiquity, lay in the Greek term eikōn which meant “image,” to refer to the ancient portrait statues of victorious athletes which were thought to bear a direct similitude with their parent divinities. For the postwar, Hollywood-film spectator, Adorno said, “the world outside is an extension of the film he has just left,” because realism is a precise instrument for the manipulation of the mass spectator by the culture industry, for which the filmic image is an advertisement for the world unedited. Mimesis, or the reproduction of reality, is a “mere reproduction of the economic base.” It is precisely film’s iconicity, then, its “realist aesthetic . . . [that] makes it inseparable from its commodity character.”...

Using conceptual tool for intuitive interaction to design intuitive website for SME in India : A case study

Statistical reports of SMEs Internet usage from various countries indicate a steady growth. However, deeper investigation of SME’s e-commerce adoption and usage reveals that a number of SMEs fail to realize the full potential of e-commerce. Factors such as lack of tools and models in Information Systems and Information Technology for SMEs, and lack of technical expertise and specialized knowledge within and outside the SME have the most effect. This study aims to address the two important factors in two steps. First, introduce the conceptual tool for intuitive interaction. Second, explain the implementation process of the conceptual tool with the help of a case study. The subject chosen for the case study is a real estate SME from India. The design and development process of the website for the real estate SME was captured in this case study and the duration of the study was four months. Results indicated specific benefits for web designers and SME business owners. Results also indicated that the conceptual tool is easy to use without the need for technical expertise and specialized knowledge.

Enhancement of relevant features for text mining

With the explosion of information resources, there is an imminent need to understand interesting text features or topics in massive text information. This thesis proposes a theoretical model to accurately weight specific text features, such as patterns and n-grams. The proposed model achieves impressive performance in two data collections, Reuters Corpus Volume 1 (RCV1) and Reuters 21578.

An analysis of theories of search and search behavior

Theories of search and search behavior can be used to glean insights and generate hypotheses about how people interact with retrieval systems. This paper examines three such theories, the long standing Information Foraging Theory, along with the more recently proposed Search Economic Theory and the Interactive Probability Ranking Principle. Our goal is to develop a model for ad-hoc topic retrieval using each approach, all within a common framework, in order to (1) determine what predictions each approach makes about search behavior, and (2) show the relationships, equivalences and differences between the approaches. While each approach takes a different perspective on modeling searcher interactions, we show that under certain assumptions, they lead to similar hypotheses regarding search behavior. Moreover, we show that the models are complementary to each other, but operate at different levels (i.e., sessions, patches and situations). We further show how the differences between the approaches lead to new insights into the theories and new models. This contribution will not only lead to further theoretical developments, but also enables practitioners to employ one of the three equivalent models depending on the data available.