Building realistic potential patients queries for medical information retrieval evaluation

摘要: To evaluate and improve medical information retrieval, benchmarking data sets need to be created. Few benchmarks have been focusing on patients’ needs. There is a for additional enable research into effective retrieval methods. In this paper we describe the manual creation of patient queries investigate their automatic generation. This work conducted in framework evaluation campaign, which aims technologies help patients laypeople access eHealth data. end, campaign composed different tasks, including (IR) task. Within IR task, web crawl medically related documents, as well are provided participants. The built represent potential needs may while reading report. We start by describing typical types then how these manually generated from reports first two years campaign. explore techniques that would us automate query generation process. This process particularly challenging, it requires an understanding needs, electronic health records. various approaches automatically generate our future development phase.