Queensland University of Technology

About: Queensland University of Technology is a education organization based out in Brisbane, Queensland, Australia. It is known for research contribution in the topics: Population & Context (language use). The organization has 14188 authors who have published 55022 publications receiving 1496237 citations. The organization is also known as: QUT.

Programme for International Student Assessment (PISA)

Abstract: The assessment of the science component of the OECD’s PISA project introduced a radically new intention for the assessment of science learning and operationalized this with a novel instrument that included item types that had not previously been used in such large-scale testing, either nationally or internationally. The OECD’s commission for the PISA project in 1998 was to provide information to participating countries about how well prepared their 15-year-old students were for twentieth-first-century life in the domains of reading, mathematics, and science – an unusually prospective brief for the assessment of learning. Fifteen-year-old students were chosen because, in a number of countries, it is the age when compulsory study of science and mathematics can cease. This commission required PISA Science to be not another retrospective assessment of students’ science learning, as is customary at the levels of classroom, school, regional, national, and international assessments (like those used by the IEA in its ongoing TIMSS project). Such testing is closely tied to the intended curriculum for science and can be used to indicate a student’s readiness to progress to the next level of schooling or to further study of the sciences beyond schooling in universities or other tertiary institutions. Future preparedness for life in society as an assessment intention was quite unknown in 1998 among the OECD countries. There were, thus, no existing models for such testing, and one had to be developed that would lead to measures of the students’ capability to apply their science knowledge to twenty-first-century contexts involving science and technology (S&T). This innovative intention to measure preparedness was applauded and endorsed by the member countries of the OECD, but there was widespread skepticism about what would be found by such a study, since the application of science knowledge in unfamiliar contexts was not something that existing science education in schools was emphasizing. It was encouraging that the students in many countries performed well on the tests although there was clear scope for improvement in all cases.

Place Recognition with ConvNet Landmarks: Viewpoint-Robust, Condition-Robust, Training-Free

Abstract: Place recognition has long been an incompletely solved problem in that all approaches involve significant compromises. Current methods address many but never all of the critical challenges of place recognition – viewpoint-invariance, condition-invariance and minimizing training requirements. Here we present an approach that adapts state-of-the-art object proposal techniques to identify potential landmarks within an image for place recognition. We use the astonishing power of convolutional neural network features to identify matching landmark proposals between images to perform place recognition over extreme appearance and viewpoint variations. Our system does not require any form of training, all components are generic enough to be used off-the-shelf. We present a range of challenging experiments in varied viewpoint and environmental conditions. We demonstrate superior performance to current state-of-the- art techniques. Furthermore, by building on existing and widely used recognition frameworks, this approach provides a highly compatible place recognition system with the potential for easy integration of other techniques such as object detection and semantic scene interpretation.

Phenomenography: A Qualitative Research Approach for Exploring Understanding in Health Care

Abstract: Phenomenography is a little-known qualitative research approach that has potential for health care research, particularly when people’s understanding of their experience is the goal. Phenomenography is explained as a qualitative, nondualistic research approach that identifies and retains the discourse of research participants. This article seeks to present the major assumptions associated with phenomenographic research. An example of the way in which research outcomes are presented is included to emphasize its distinctiveness. It is noted that phenomenography has potential in the area of qualitative health research and will benefit from ongoing development and application.

Whole-genome sequencing reveals untapped genetic potential in Africa’s indigenous cereal crop sorghum

Abstract: Sorghum is a food and feed cereal crop adapted to heat and drought and a staple for 500 million of the world’s poorest people. Its small diploid genome and phenotypic diversity make it an ideal C4 grass model as a complement to C3 rice. Here we present high coverage (16-45 × ) resequenced genomes of 44 sorghum lines representing the primary gene pool and spanning dimensions of geographic origin, end-use and taxonomic group. We also report the first resequenced genome of S. propinquum, identifying 8 M high-quality SNPs, 1.9 M indels and specific gene loss and gain events in S. bicolor. We observe strong racial structure and a complex domestication history involving at least two distinct domestication events. These assembled genomes enable the leveraging of existing cereal functional genomics data against the novel diversity available in sorghum, providing an unmatched resource for the genetic improvement of sorghum and other grass species.

Predictive Business Process Monitoring with LSTM Neural Networks

Abstract: Predictive business process monitoring methods exploit logs of completed cases of a process in order to make predictions about running cases thereof. Existing methods in this space are tailor-made for specific prediction tasks. Moreover, their relative accuracy is highly sensitive to the dataset at hand, thus requiring users to engage in trial-and-error and tuning when applying them in a specific setting. This paper investigates Long Short-Term Memory (LSTM) neural networks as an approach to build consistently accurate models for a wide range of predictive process monitoring tasks. First, we show that LSTMs outperform existing techniques to predict the next event of a running case and its timestamp. Next, we show how to use models for predicting the next task in order to predict the full continuation of a running case. Finally, we apply the same approach to predict the remaining time, and show that this approach outperforms existing tailor-made methods.