Multi-Ontology Learning: Crossing the Boundaries of Domains and Languages
Final Report Abstract
The main focus of the MULTIPLA project was on the development of methods that allow to generate, to improve and to apply multilingual knowledge structures, ontologies in particular. One important goal was to develop methods that can exploit these multilingual knowledge structures for the task of cross-lingual information retrieval. Several achievements have been reached in the MULTIPLA project. First, we showed that user-generated content from popular knowledge and community portals can be used as multilingual resource. This allows to exploit the growing Web 2.0 consisting of millions of contributions by users from various linguistic regions. As an important contribution, we developed new approaches to build knowledge structures from these multilingual resources and we showed the actual benefit of exploiting such structures in the conlext of multilingual search and retrieval tasks. Second, we introduced several methods to improve the quality of multilingual resources. This includes approaches to ontology alignment and to ontology learning. In this line of research, we also investigated in analyzing the quality of existing Web 2.0 datasets. Third, research investigations in the field of Web mining have emphasized the role of multi-aspect content analysis for knowledge extraction tasks from Social Media. In the context of the MULTIPLA project we organized two international workshops on multilingual expert search and one on detecting and exploiting cultural diversity on the Social Web. Further, we established a multilingual retrieval challenge for multilingual expert search on a Web 2.0 dataset. These successful events were part of the project dissemination work and showed the interest of other groups on the research in MULTIPLA. The research results from the MULTIPLA project have been applied in several other projects at AIFB and WeST, such as RENDER., Active, Live+Gov, and SocialSensor. This shows that the foundational research results that were achieved in this project can be exploited in different application scenarios.
Publications
- An Experimental Comparison of Explicit Semantic Analysis Implementations for Cross-Language Retrieval. In: Proceedings of the International Conference on Applications of Natural Language to Information Systems (NLDB), Saarbruecken, Germany, 2009: 36-48
Philipp Sorg, Philipp Cimiano
- Explicit vs. Latent Concept Models for Cross-Language Information Retrieval. In: Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), Pasadena, USA, 2009: 1513-1518
Philipp Cimiano, Antje Schulz, Sergej Sizov, Philipp Sorg, Steffen Staab
- Natural Language Interfaces: What is the Problem? - A data-driven quantitative analysis. In: Proceedings of the International Conference on Applications of Natura Language to Information Systems (NLDB), Saarbrücken, Gerrnany, 2009: 192-200
Philipp Cimiano, Michael Minock
- TripleRank: Ranking Semantic Web Data by Tensor Decomposition. In: Proceedings of ISWC-2009 International Semantic Web Conference (ISWC 2009), Westfield, USA, 2009: 213-228
Thomas Franz, Antje Schultz, Sergej Sizov, Steffen Staab
- An Unsupervised Approach for Acquiring Ontologies and RDF Data from Online Life Science Databases. In: 7th Extended Semantic Web Conference (ES-WC2010), Heraklion, Greece, 2010: 310-333
Saqib Mir, Steffen Staab, Isabel Rojas
- Analysing Dependency Dynamics in Web Data. In: Proceedings of Linked AI: AAAI Spring Symposium ''Linked Data Meets Artificial Intelligence", Stanford, USA, March 2010
Felix Biemann, Andreas Harth
- Compositional Matrix-Space Models of Language. In: Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, Uppsala, Sweden, 2010: 907-916
Sebastian Rudolph, Eugenie Giesbrecht
- DL-Link: A conceptual clustering algorithm for indexing description logics knowledge bases. International Journal of Semantic Computing, 4(4): 453-486, 2010
Claudia d'Amato, Steffen Staab, Nicola Fanizzi, Floriana Esposito
- GeoFolk: latent spatial semantics in web 2.0 social media. In: Proceedings of the Third International Conference on Web Search and Web Data Mining, WSDM 2010, New York, USA, 2010: 281-290
Sergej Sizov
- SXPath Extending XPath towards Spatial Querying on Web Documents. In: PVLDD Proceedings of the Very Large Databases Endowment, 4(2):129-140, 2010
Linda dOro, Massimo Ruffolo, S. Staab
- Discriminating graphs through spectral projections. Computer Networks 55(15):345S-3468 (2011)
Damien Fay, Hamed Haddadi, Steve Uhlig, Liam Kilmartin, Andrew W. Moore, Jerome Kunegis, Marios Iliofotou
- Finding the Right Expert: Discriminative Models for Expert Retrieval. In: Proceedings of the International Conference on Knowledge Discovery und Information Retrieval (KDIR), Paris, France, 2011
Philipp Sorg, Philipp Cimiano
- Language resources extracted from Wikipedia. In: Proceedings of the International Conference on Knowledge Capture (K-CAP), Banff, Canada, 2011: 153-160
Denny Vrandecic, Philipp Sorg, Rudi Studer
- Lexinfo: A declarative model for the lexicon-ontology interface. J. Web Sem. (WS) 9(1):29-51 (2011)
Philipp Cimiano, Paul Buitelaar, John McCrae, Michael Sintek
- Latent Geospatial Semantics of Social Media. ACM Transactions on Intelligent Systems and Technology (TIST), 2012
Sergej Sizov
