Northumbria University

About: Northumbria University is a education organization based out in Newcastle upon Tyne, United Kingdom. It is known for research contribution in the topics: Context (language use) & Population. The organization has 5624 authors who have published 17423 publications receiving 381949 citations. The organization is also known as: University of Northumbria at Newcastle.

The human touch: participatory practice and the role of facilitation in designing with communities

Abstract: Traditional PD research offers a range of methods for participant engagement. Yet little is shared of the microdynamics of participation at its most intense, when the designer as facilitator is challenged by a range of social contingencies. Engaging people in change can be a messy process, especially when emotions run high. This paper explores two situations where communities were asked to collaborate on disaster mitigation plans and looks at how facilitation took place to engage with these concerns. It considers the relationship between method and its enactment, between the participatory practitioner and participant group, and between intention and outcome. In doing so, it questions the prevalent research culture that anonymises facilitation and its agency. Instead, we offer a more synthesised reading of practice, by focusing on aspects that compound the designers' task, such as the dynamics of the group and emotions manifested by participants. We argue that we need to orientate towards understanding the designers' participatory practice, rather than reporting participatory methods alone. The act of engaging others involves an embodied knowing, with moment-by-moment shifts in position, focus and delivery. Acknowledging this involves a rethink of our frameworks for reflecting and reporting on design.

Past East Asian monsoon evolution controlled by paleogeography, not CO2

Abstract: The East Asian monsoon plays an integral role in human society, yet its geological history and controlling processes are poorly understood. Using a general circulation model and geological data, we explore the drivers controlling the evolution of the monsoon system over the past 150 million years. In contrast to previous work, we find that the monsoon is controlled primarily by changes in paleogeography, with little influence from atmospheric CO2. We associate increased precipitation since the Late Cretaceous with the gradual uplift of the Himalayan-Tibetan region, transitioning from an ITCZ-dominated monsoon to a sea breeze–dominated monsoon. The rising region acted as a mechanical barrier to cold and dry continental air advecting into the region, leading to increasing influence of moist air from the Indian Ocean/South China Sea. We show that, apart from a dry period in the middle Cretaceous, a monsoon system has existed in East Asia since at least the Early Cretaceous.

Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing

Abstract: For efficiently retrieving nearest neighbors from large-scale multiview data, recently hashing methods are widely investigated, which can substantially improve query speeds. In this paper, we propose an effective probability-based semantics-preserving hashing (SePH) method to tackle the problem of cross-view retrieval. Considering the semantic consistency between views, SePH generates one unified hash code for all observed views of any instance. For training, SePH first transforms the given semantic affinities of training data into a probability distribution, and aims to approximate it with another one in Hamming space, via minimizing their Kullback–Leibler divergence. Specifically, the latter probability distribution is derived from all pair-wise Hamming distances between to-be-learnt hash codes of the training data. Then with learnt hash codes, any kind of predictive models like linear ridge regression, logistic regression, or kernel logistic regression, can be learnt as hash functions in each view for projecting the corresponding view-specific features into hash codes. As for out-of-sample extension, given any unseen instance, the learnt hash functions in its observed views can predict view-specific hash codes. Then by deriving or estimating the corresponding output probabilities with respect to the predicted view-specific hash codes, a novel probabilistic approach is further proposed to utilize them for determining a unified hash code. To evaluate the proposed SePH, we conduct extensive experiments on diverse benchmark datasets, and the experimental results demonstrate that SePH is reasonable and effective.