In this software project, we deal with the challenge of heterogeneous data on the Web. Often, we cannot answer a query from a single source and need to compose information from multiple information sources that use different terminologies to describe the objects that generate heterogeneity problem e.g., the terms ‘PersonName’ and ‘person-name’ in two different sources refer to the same object. Ontologies are widely used to share the information but they tend to be highly heterogeneous. Moreover, they are modeled with rich object and data properties, rather than gleaned with majorly hierarchical relationship. This richer representation demands for sophisticated tools for aligning these ontologies. We leverage this fact and develop a software that provides mapping or alignment between such heterogeneous ontologies. As a practical use case, we employ ontologies from the Terminology Service (https://terminologies.gfbio.org) of German Federation for Biological Data (GFBio) project which is a sustainable, service oriented, national data infrastructure facilitating data sharing for biological and environmental research.
All the ontologies of GF-Bio project are available in OWL and SKOS languages. We will provide syntactic mapping (mapping according to the syntax) between the concepts of the ontologies and evaluate the results in terms of precision, recall and F-measure. We will also conduct subjective testing to gain a deep intuitive understanding of the performance of the proposed approach. The tasks of this project are as follows
1. Create mapping between ontologies
2. Evaluate the mapping results
The main task of the students is to learn and code the algorithms that are given by the mentor (Archana Patel). Mentor will always be ready to help the students to resolve their problems.
,