Humanities and Social Sciences – SDU eScience http://escience.sdu.dk Mon, 02 Dec 2019 08:52:38 +0000 en-US hourly 1 https://wordpress.org/?v=5.5.20 The advantages of using a supercomputer at SDU http://escience.sdu.dk/index.php/2019/12/02/the-advantages-of-using-a-supercomputer-at-sdu/ Mon, 02 Dec 2019 08:34:13 +0000 http://escience.sdu.dk/?p=5035 Read more…]]> Have you ever wondered how a supercomputer works and what it can do for your research or teaching?

At the eScience Center we can provide the support you need to solve large, advanced, complex computational problems which are too big for a single machine to compute.

Check out the below videos for more information.

]]>
Automatic generation of rap lyrics http://escience.sdu.dk/index.php/2018/04/05/automatic-generation-of-rap-lyrics/ Thu, 05 Apr 2018 11:55:06 +0000 http://35.198.183.193/?p=3834 Read more…]]> Principal Investigator: Manex Aguirrezabal Zabaleta

Centre for Language Technology – Department of Nordic Studies and Linguistics – KU

We are interested in automatically generating rap lyrics, using a dataset with over 30,000 rap songs. Past approaches in poetry/verse generation typically involve analytical approaches with hand-crafted grammars and rules. In rap lyrics generation, there are hybrid approaches, where the authors devise an architecture to structurally analyze rap texts and suggest the closest matching next line from a database. On the other hand, we have pure unsupervised approaches using the Long Short Term Memory (LSTM) Recurrent neural networks (RNN) variant with promising results, albeit using a limited corpus. Recently, Generative Adversarial Networks (GANs) and similar methods have been proved to model well specific styles.

]]>
Digital Research Infrastructures and eScience in Archaeology & Heritage http://escience.sdu.dk/index.php/2017/11/01/digital-research-infrastructures-and-escience-in-archaeology-heritage/ Wed, 01 Nov 2017 14:00:11 +0000 http://35.198.183.193/?p=555 Read more…]]> Principal Investigator: Associate Professor Jens-Bjørn Riis Andresen

School of Culture and Society – Department of Archeology and Heritage Studies – Aarhus University (AU)

Archaeological data, information, and knowledge is increasingly recorded, stored, managed, processed, and disseminated on digital media as its primary carrier. This development constitutes both a challenge and opens new avenues of possibilities for the discipline. Only through inter-disciplinary corporation satisfactory solutions can be accomplished. We participate in initiatives on both EU (http://www.nedimah.eu/; http://www.dariah.eu/; http://www.ariadne-infrastructure.eu/; https://www.egi.eu/news/eosc-hub-project-is-favourably-evaluated/) and national level (DIGHUMLAB), where research initiatives have been launched to facilitate seamless work-flows and long-term storage solutions for digital data. We focus on space and time modelling, dimensions which are of central importance in archaeological reasoning. Other key research areas are: 3D visualization & modelling, remote sensing, predictive modelling, digital heritage management, excavation recording, the semantic web, user / community participation. We have recently created a national section of the CAA and have organized the first Nordic conference in May (http://conferences.au.dk/digdata2017/).

HPC is important for our research because some of the applications are time-critical and one needs a computer with necessary power to get the job done in time. The data, we are running are primarily applications for digital field recording – computer vision applications.

Our results so far is described in the paper “Supercomputing at the trench edge”. The paper is accepted for the global CAA-proceedings Oslo 2016.

]]>