The upcoming International Conference on Theory and Practice of Digital Libraries (TPDL) will take place from Monday, September 10, to Thursday, September 13, 2018, at the Faculty of Engineering of the University of Porto (FEUP) in Portugal. The TPDL conference started in 1997 as the European Conference on Research and Advanced Technology in Digital Libraries (ECDL), laying thus the foundation for the establishment of a European Research Community in the domain of Digital Libraries. The conference has become a notable and prestigious European forum, focusing on digital libraries and all associated technical, practical, and social aspects.
The following TIB publications and posters were accepted at TPDL:
- "Unveiling Scholarly Communities over Knowledge Graphs", by Sahar Vahdati, Guillermo Palma, Rahul Jyoti Nath, Christoph Lange, Sören Auer and Maria-Esther Vidal.
Info: In this work, the bibliographic metadata are integrated from various sources and analysed in order to identify common collaboration structures. The work constitutes an important prerequisite for TIB’s strategic goal of developing the Open Research Knowledge Graph.
- "Metadata Analysis of Scholarly Events of Computer Science, Physics, Engineering, and Mathematics", by Said Fathalla, Sahar Vahdati, Sören Auer and Christoph Lange.
Info: This work analyses the evolution of conferences and respective publications in various scientific and technological fields, shedding thus light on the changes caused by the digitization of scholarly communication and its publishing over the last decades. (preprint)
- "Figures in Scientific Open Access Publications", by Lucia Sohmen, Jean Charbonnier, Ina Blümel, Christian Wartena and Lambert Heller.
Info: In this work, a statistical analysis of the image corpus is published, which is derived from the collection of over one million open access articles and their respective extracted images. The objective of this publication is to give an overview of the content of open access publications, demonstrating thus the requirements in order to establish a corpus of more than a million pictures from the Open Access literature, which will be searchable and reusable within the context of Wikimedia, and respectively Wikipedia.
- "Finding Person Relations in Image Data of News Collections in the Internet Archive", by Eric Müller-Budack, Kader Pustu-Iren, Sebastian Diering and Ralph Ewerth.
Info: In this paper, an automatic deep learning based system for person identification in image data of news collections of the Internet Archive is introduced. An appropriate visualization of the results allows users and researchers, e.g. historians, to quickly retrieve information about relationships of entities, i.e., how often persons of public interest have been portrayed in the same photo during a certain period of time.
- "Recommending Scientific Videos based on Metadata Enrichment using Linked Open Data", by Justyna Medrek, Christian Otto and Ralph Ewerth.
Info: This paper presents a novel video recommendation approach based on linked open data for the TIB AV-Portal. The recommendation of semantically similar videos is based on two types of information, while automatically extracted metadata for every video are enriched with information from the Integrated Authority File (GND) of the German National Library. A user study demonstrates the feasibility of the approach.
- "TIB-arXiv: An Alternative Search Portal for the arXiv Pre-print Server", by Matthias Springstein, Huu Hung Nguyen, Anett Hoppe and Ralph Ewerth.
Info: The original Web portal of the popular open access pre-print server arXiv provides basic search functionality. In this paper, we present a novel user interface (https://labs.tib.eu/arxiv) with additional features in order to improve the search functionality in the entire set of arXiv papers.
- "An Analytics Tool for Exploring Scientific Software and Related Publications", by Anett Hoppe, Jascha Hagen, Helge Holzmann, Günter Kniesel and Ralph Ewerth.
Info: Scientific software constitutes a key element for reproducible research. However, classic publications and related scientific software are typically not (sufficiently) linked, lacking tools that allow users to jointly explore these artefacts. In this paper, we report on our work on developing an analytics tool, called SciSoftX, for jointly exploring software and publications. The presented prototype and the resepctive two use cases demonstrate the usefulness of the approach.