Name of the participant: Andreas Thalhammer
Description of the IT research project: The SumOn project was concerned with the development of a flexible system that enables automatic summaries of linked data entities. The focus was on resolving the classic type-related presentation of entities as well as the inclusion of various context factors. These factors included an orientation regarding importance (normal everyday thinking), the reference to current events and a language-oriented component. The SumOn system was implemented with various application and user interfaces, for example the “SUMMA API” and “ELES”, a demo application which interactively combines “Entity Linking” and “Entity Summarization”:
- Which data sources can best support the factors of importance, timeliness and language context with regard to entity summarization?
- Which algorithms and methods of machine learning are suitable for ranking facts about individual entities?
- How are entity summarization, visualization and navigation combined to make the best possible use of synergies?
- Which methods can be used to automatically merge entity related data from different knowledge bases on the net?
The results of the SumOn project led to three conference publications, two workshop publications and a demo presentation. Two of these works were awarded prizes, and another was nominated. In addition to the scientific publications, various data sets were also published. In particular, the data sets “Wikidata PageRank” and “DBpedia PageRank” were used in a variety of ways in further research work. A large part of the research results obtained from SumOn formed the core of the dissertation “Linked Data Entity Summarization”, which was completed in 2016.
Software Campus partners: Karlsruher Institut für Technologie, SAP SE
Implementation period: 01.03.2015 – 28.02.2017
Publications:
Andreas Thalhammer, Steffen Thoma, Andreas Harth, and Rudi Studer. Entity-centric Data Fusion on the Web. Proceedings of the 28th ACM Conference on Hypertext and Social Media, HT ‘17, pages 25-34. ACM, New York, NY, USA, 2017.
*ACM SIGWEB Ted Nelson Newcomer Award*
Andreas Thalhammer and Steffen Stadtmüller. SUMMA: A Common API for Linked Data Entity Summaries. In Engineering the Web in the Big Data Era, volume 9114 of Lecture Notes in Computer Science, pages 430–446. Springer International Publishing, Cham, 2015.
*Best Research Paper Nominee*
Weitere Informationen: https://entitysummarization.org