FEVER - Framework for EValuating Entity Resolution

FEVER is a new evaluation platform for entity resolution approaches. The modular structure of the FEVER framework supports the incorporation or reconstruction of many previously proposed approaches for entity resolution. A distinctive feature of FEVER is that it not only evaluates traditional measures such as precision and recall but also the effort for configuring (e.g., parameter tuning, training) a good entity resolution approach.

Architecture

The above figure illustrates the architecture of FEVER consisting of modules for the definition and execution of entity resolution approaches or so-called match workflows. A GUI-based workflow editor allows to interactively specify match workflows utilizing operator trees. The FEVER runtime executes a specified workflow and, thus, generates evaluation data.

Comparative evaluation of entity resolution approaches with FEVER

Analysis of entity resolution publications

For our comparative evaluation of entity resolution approaches in [1] we have analyzed numerous research publications w.r.t. their evaluation of entity resolution approaches. The complete list can be found here.

Datasets

The datasets used in the evaluation can be downloaded here

Publications

Köpcke, H.; Thor, A.; Rahm, E. Evaluation of entity resolution approaches on real-world match problems Proc. 36th Intl. Conference on Very Large Databases (VLDB) / Proceedings of the VLDB Endowment 3(1), 2010 2010-09

Thor, A. Toward an adaptive String Similarity Measure for Matching Product Offers Proc. GI-Workshop - Informationsintegration in Service-Architekturen, 2010 2010-09

Köpcke, H.; Thor, A.; Rahm, E. Learning-based approaches for matching web data entities IEEE Internet Computing 14(4), 2010 2010-07

Köpcke, H.; Thor, A.; Rahm, E. Comparative evaluation of entity resolution approaches with FEVER Proc. 35th Intl. Conference on Very Large Databases (VLDB), 2009 (demo) 2009-08

Contact/Project Members

• Prof. Dr. Erhard Rahm
• Hanna Köpcke
• Dr. Andreas Thor