Boosting the Efficiency of Large-Scale Entity Resolution with Enhanced Meta-Blocking

Entity Resolution constitutes a quadratic task that typically scales to large entity collections through blocking. The resulting blocks can be restructured by Meta-blocking to raise precision at a limited cost in recall. At the core of this procedure lies the blocking graph, where the nodes correspond to entities and the edges connect the comparable pairs. There are several configurations for Meta-blocking, but no hints on best practices. In general, the node-centric approaches are more robust and suitable for a series of applications, but suffer from low precision, due to the large number of unnecessary comparisons they retain. In this work, we present three novel methods for nodecentric Meta-blocking that significantly improve precision. We also introduce a pre-processing method that restricts the size of the blocking graph by removing a large number of noisy edges. As a result, it reduces the overhead time of Meta-blocking by 2 to 5 times, while increasing precision by up to an order of magnitude for a minor cost in recall. The same technique can be applied as graph-free Meta-blocking, enabling for the first time Entity Resolution over very large datasets even on commodity hardware. We evaluate our approaches through an extensive experimental study over 19 voluminous, established datasets. The outcomes indicate best practices for the configuration of Meta-blocking and verify that our techniques reduce the resolution time of state-ofthe-art methods by up to an order of magnitude.