BENGAL: An Automatic Benchmark Generator for Entity Recognition and Linking
This addresses the resource-intensive and error-prone nature of benchmark creation for entity recognition and linking, offering an incremental improvement by automating the process.
The authors tackled the problem of manual creation of error-prone gold standards for named entity recognition and linking by developing BENGAL, an automatic benchmark generator that produces cost-effective, error-free benchmarks, showing it can create varied benchmarks with characteristics similar to existing ones across multiple languages.
The manual creation of gold standards for named entity recognition and entity linking is time- and resource-intensive. Moreover, recent works show that such gold standards contain a large proportion of mistakes in addition to being difficult to maintain. We hence present BENGAL, a novel automatic generation of such gold standards as a complement to manually created benchmarks. The main advantage of our benchmarks is that they can be readily generated at any time. They are also cost-effective while being guaranteed to be free of annotation errors. We compare the performance of 11 tools on benchmarks in English generated by BENGAL and on 16benchmarks created manually. We show that our approach can be ported easily across languages by presenting results achieved by 4 tools on both Brazilian Portuguese and Spanish. Overall, our results suggest that our automatic benchmark generation approach can create varied benchmarks that have characteristics similar to those of existing benchmarks. Our approach is open-source. Our experimental results are available at http://faturl.com/bengalexpinlg and the code at https://github.com/dice-group/BENGAL.