Homepage of Ansgar Scherp: About Me (http://ansgarscherp.net/#me)

Menu: [Overview] • [About Me] • [Projects] • [Publications] • [Dissertation] • [Habilitation]


Education and Professional Experience

Image showing a Portrait of Ansgar Scherp Ansgar Scherp is Professor for Knowledge Discovery with the Institute of Computer Science, Faculty of Engineering at
Kiel University, Germany and ZBW – Leibniz Information Centre for Economics (short: ZBW) in Kiel since January 2014. Ansgar Scherp was Juniorprofessor for Media Informatics and was member of the Research Group on Data and Web Science of the University of Mannheim, Germany from August 2012 to December 2013. Since April 2013, he was also associated professor with the Institute for Enterprise Systems (InES) in Mannheim. Prior to that he worked as Juniorprofessor for Semantic Web with the University of Koblenz-Landau in the Institute for Information Systems Research since April 2011 and lead the focus group on Interactive and Multimedia Web at the Institute for Web Science and Technologies (WeST) at the same university since May 2008. He has studied computer science at the University of Oldenburg, Germany and has received the Advancement Award for Outstanding Results in Studies from the Association for Electrical, Electronic & Information Technologies (VDE), Germany in 1998. He finished his PhD with the thesis title "A Component Framework for Personalized Multimedia Applications" at the University of Oldenburg, Germany with distinction in 2006. Mr. Scherp has been EU Marie Skłodowska Curie Fellow with Prof. Ramesh Jain at the Donald Bren School of Information and Computer Sciences, University of California, Irvine, USA in Los Angeles between November 2006 to October 2007. Subsequently, he has lead the University of Koblenz-Landau's activities in the EU Integrated Project WeKnowIt from 2008 to 2011. Here, he has been leading the work packages on knowledge management and mass intelligence and has been member of the project management board and steering board committee. Mr. Scherp has been scientific leader of the EU project SocialSensor, where the University of Koblenz-Landau lead the work package on user modeling and presentation. Currently, he is scientific leader of the EU H2020 project MOVING on training data-savvy information professionals. In December 2011, he has received his Venia Legendi (Habilitation) with the thesis title "Semantic Media Management: Process Innovation along the Value Chain of Media Companies" (in German) from the University of Koblenz-Landau, Germany. He has published over 60 peer-reviewed scientific publications including more than 20 journal articles, 50 conference papers, and 10 book chapters.

Research Interests

Currently, I am scientific leader of the large-scale interdisciplinary EU H2020 project
MOVING running from April 2016 to March 2019 on enabling young researchers, decision makers, and public administrators to employ and use data mining tools to search, organize, and manage large-scale information sources. The MOVING platform integrates these tools in a common system and provides new ways for semantic search over large corpora of scientific publications, websites, social media, and videos. In addition, it provides novel graph-based visualizations to explore the content beyond standard list-based results of web search engines.

The MOVING project is a good and representative example of what motivates me and my research. I conduct research on novel methods for analyzing very-large datasets of unstructured, textual content for tasks such as document classification [C46, C60] and recommendation [C55]. Furthermore, I am developing efficient data mining approaches for analyzing very-large, distributed graph data on the web, the so-called Linked Open Data (LOD) cloud. I have developed SchemEX, an approach for the efficient stream-based computation of a schema-level index of very-large LOD graphs under the constraint of real-time processing [J12]. SchemEX can extract the schema information from the LOD cloud in linear runtime with respect to the number of input data, i. e., graph edges. A schema-level index like SchemEX allows users in finding and exploring data sources based on queries that are formulated from combinations of RDF types and/or properties. Internally, the index is represented as a graph, i. e., the SchemEX index can also be considered as kind of meta-knowledge of LOD. SchemEX allows to answer queries to find distributed data sources on the web, like Google does this for web documents. The idea of a stream-based schema computation in SchemEX won the Billion Triple Challenge of the International Semantic Web Conference in 2011. The data search engine LODatio2 allows end users to interactively search and explore for relevant sources of data by using the SchemEX index („yellow pages“ for LOD) [C27].

In addition, I have conducted extensive analyses of schema structures of LOD on different levels of granularity and use of information-theoretic measures such as mutual information [J17, C24]. Furthermore, I have applied formal concept analysis on LOD and have extended formal concept analysis by the notion of parallel lattices [C35]. A parallel lattice constructs multiple lattices from the same data set but over different partitions or attributes at the same time. This allows to navigate between different lattices and use the parallel lattices for scenarios such as query recommendations.

Furthermore, I am using machine learning methods like Learning to Rank to support data engineers in modeling the schema of their Linked Open Data [C54]. Currently, I am investigating the evolution of LOD, i. e., the temporal dynamics of very-large graphs on the web. Goal is to develop strategies for an efficient crawling of such graph data and to keep data caches up-to-date [C58, C45]. Selected research results are published as Open Source and thus made available to the research community. You find an overview of the Open Source tools published so far on the homepage of the working group under Tools and Source Code.

Beyond this, I have strong experience in data analysis and use of machine learning methods for human-computer interaction, particularly in the area of applying data analysis to eye-tracking data and using it in applications like photo search [C33] or personalized photo selection [C34]. In general, I demonstrate a strong commitment to user-centered development and evaluation of the resulting applications using quantitative, i. e., data-driven and statistical analyses techniques. One example is the mobile, location-based application mobEx [C51, C41]. The location-based application allows for integrated search for places and events over multiple data providers. Specific feature of mobEx is that it integrates query responses from different social media data providers and merges them in real-time using an efficient entity resolution engine [C51]. At the same time, the application employs a novel user interface for an integrated map-based and time-based exploration of the social media entities using the notion of a time-slider that is integrated with Google Maps. The idea of mobEx has won the KlickTel Award4 with a price money of 10.000 € in 2013.

Honors and Awards

Community Service

Mr. Scherp is editor of the Journal of Web Semantics (JWS) since 2010. He is program committee member for conferences including World Wide Web (WWW), ACM Multimedia (MM), Multimedia Modeling (MMM), Extended Semantic Web Conference (ESWC), and International Semantic Web Conference (ISWC). He also reviews for journals including Proceedings of Very Large Data Base Endowment (PVLDB), IEEE Multimedia, Springer's Multimedia Systems and Multimedia Tools and Applications (MTAP), ACM Transactions on Multimedia Computing Communications and Applications (TOMCCAP), Journal of Web Semantics (JWS), and International Journal on Human Computer Studies (IJHCS). Mr. Scherp is co-organizer of several scientific events such as the ACM Workshop series on Events in Multimedia conjunct with ACM Multimedia Beijing, China, 2009 and Firence, Italy, 2010, and Scottsdale, AZ, USA, in 2011. The workshop aims at bringing together different disciplines interested in detecting, processing, representing, and using events in multimedia and social media. Due to the workshop's success, the topic became its own area at the ACM Multimedia conference in 2012. Furthermore, Mr. Scherp led the doctoral programme of INFORMATIK, the annual German computer science society meeting, between 2013 and 2015.


Keynote Talks

Invited Talks

Most Important Publications

For a complete list, please refer to the list here or to my DBLP page.

Supervised Phd Theses

Supervised Master Theses

(incomplete list)

Last update: 6/12/2018.