Master or Diploma Thesis: Automatic Discovery and Interlinking of Dataset Resources in the Linked Open Data Cloud

Title:
Automatic Discovery and Interlinking of Dataset Resources in the Linked Open Data Cloud

Description:
The number of datasets in the Linked Open Data Cloud increases rapidly and has reached a total of 295 by September 2011. These datasets provide a huge pool of contextual information interlinkable to each other that can be exploited by context-aware services. For example, location information coming from the GeoNames dataset can be interlinked to a certain point-of-interest (POI) in the DBPedia dataset providing detailed information about this POI.

However, due to the huge amount of data present in the LOD Cloud, finding suitable datasets providing useful contextual resources to interlink newly created resources with is very time-consuming and requires manual effort. Furthermore, there is no concrete evaluation schema that can be taken into consideration in order to interlink the "best" and "most popular" resources.

The main objective of this thesis is to develop concepts and solutions for automatic discovery and interlinking of contextual resources in the Linked Open Data Cloud with newly created dataset resources. Furthermore, an evaluation schema has to be worked out, which guarantees that only the "best" and "most popular" datasets are chosen for the interlinking process.

Challenges:

• Analyze the principles of Linked Data
• Create concepts and solutions for automatic discovery and interlinking of contextual resources in the Linked Open Data Cloud
• Develop an evaluation schema for datasets, so that only the "best" and "most popular" resources are taken into consideration
• Demonstrate the concepts and solutions via an exemplary dataset

Prerequisites:

• Interest in Linked Data Principles
• Knowledge of Semantic Web might be useful
• Proficiency in object-oriented programming

Supervisor: Prof. Dr. Axel Küpper, Abdulbaki Uzun

Type: Master or Diploma Thesis

Duration: 6 months