OGER: OntoGene’s Entity Recogniser in the BeCalm TIPS Task

Abstract

We present OGER, an annotation service built on top of OntoGene’s biomedical entity recognition system, which participates in the TIPS task (technical interoperability and performance of annotation servers) of the BeCalm (biomedical annotation metaserver) challenge. The annotation server is a web application tailored to the needs of the task, using an existing biomedical entity recognition suite. The core annotation module uses a knowledge-based strategy for term matching and entity linking. The server’s architecture allows parallel processing of annotation requests for an arbitrary number of documents from mixed sources. In the discussion, we show that network latency is responsible for significant overhead in the measurement of processing time. We compare the preliminary key performance indicators with an analysis drawn from the server’s log messages. We conclude that our annotation server is ready for the upcoming phases of the TIPS task.