Ontology engineering - Ontology Alignment. (PI: Patrick Lambrix)
Ontologies are a key technology for the Semantic Web. Many ontologies have been developed in the last decade and many of these contain overlapping information. Ontology alignment deals with finding mappings beween terms in different ontologies. Our work includes; development of a framework and tool (SAMBO, Winner of the Anatomy track of the 2008 Ontology Alignment Evaluation Initiative ) for aligning and merging ontologies, development of a framework and tool (KitAMO) for evaluation and comparison of ontology alignment strategies, and development of framework for recommending ontology alignment strategies. In the recent years we have focused on the challenge of supporting user involvement through the proposal of requirements for ontology alignment systems and the development of a session-based framework and system.
- SAMBO - aligning and merging ontologies
- KitAMO - tookit for evaluating ontology alignment strategies
Ontology engineering - Ontology Debugging and Completion. (PI: Patrick Lambrix)
Developing ontologies is not an easy task and often the resulting ontologies are not consistent or complete. Such ontologies, although often useful, lead to problems when used in semantically-enabled applications. Wrong conclusions may be derived or valid conclusions may be missed. To deal with this problem we may want to debug and complete the ontologies. In our work we have proposed the first method and system for repairing missing is-a structure of ontologies and an integrated framework for debugging and alignmnet of ontologies. We also co-founded the Workshop on Debugging Ontologies and Ontology Mappings and its descendant the Workshop on Completing and Debugging the Semantic Web.
- RepOSE (Repairing of Ontological Structure Environment).
Learning graphical models of and dependencies in gene networks. (PI: José M. Peña)
Most if not all diseases are not determined by a single factor, but rather by a network of interacting genetic and environmental factors. Knowledge about such a network would allow to develop better early diagnostics and drugs. The challenge is now to develop sound methodologies for learning the referred network from these measurements. An analogous challenge appears in fault diagnosis. For instance, the technique can also be used for fault diagnosis for Scania trucks.
- Learning Probabilistic Graphical Models of Gene Networks and Fault Networks
- Learning Bayesian network models of the atherosclerosis network
Efficient data management for the web (PI: Lena Strömbäck)
The Web have enabled timely and efficient access to a variety of data in different domains, and increasingly, these data are represented and exchanged in XML. XML is a flexible exchange format that can represent many classes of data. However, supporting the flexibility that makes XML attractive to different applications is challenging and there is a growing need for efficient and reliable XML data management tools and techniques. The goal for this work is to explore and develop such tools.
- Efficient analysis and management of XML data
- Hybrid XML storage mappings for efficient data management on the Web
Share and reuse of scientific workflows (PI: Juliana Freire and Lena Strömbäck)
Scientific workflows and workflow based systems have emerged as an alternative to ad-hoc approaches for documenting computational experiments and design complex processes. These workflows are complex objects and it is an important issue for the research community how to build, share and reuse them efficiently. This work aims at enabling reuse of scientific workflows through improving the usability of workflow search engines and the use of a common provenance model to integrate data provenance.
- Search engines and result presentation for scientific workflows
- Enabling interoperability and co-work for scientific workflows.
Gene normalization in biomedical text mining (PI: He Tan)
Gene normalization is to link terms appear in the texts to their entries in biological databases. The task is challenging even for domain experts, since there is no community wide agreement on how a particular gene and gene product should be named. Traditional word sense disambiguation methods simply do not deliver in a complex sublanguage such as that of biomedicine. This project is to develop a framework and methods for gene normalization based on domain knowledge obtained from ontologies and biological databases.
Similarity-based grouping of biological data (PI: Patrick Lambrix)
Development of method for similarity-based grouping of data entries. Development of tool for evaluation of similarity-based grouping strategies (KitEGA).
- KitEGA - tookit for evaluating grouping algorithms
Integration of data sources (PI: Patrick Lambrix)
Definiton of requirements for information integration systems for biomedical databases. Development of small prototype integration system (BioTRIFU). Identification of use of ontology-based information for information integration. Development of strategies for identifying similar data entries.
- BioTRIFU - transparent access to multiple biological data sources
- KitEGA - tookit for evaluating grouping algorithms
Page responsible: Patrick Lambrix
Last updated: 2016-09-27