CLEFeHealth 2014 Task 3 Evaluation Package

Full Official Name: CLEFeHealth 2014 Task 3 Evaluation Package
Submission date: Feb. 26, 2015, 12:55 p.m.

The CLEF Initiative (Conference and Labs of the Evaluation Forum) promotes the systematic evaluation of information access systems through experimentation on shared tasks, with an emphasis on multilingual and multimodal information. It is structured in two main parts: a series of Evaluation Labs for the systems devised and a peer-reviewed conference on a broad range of issues related to the project. The CLEFeHealth 2014 Task 3 Evaluation Package contains data used for the User-centred health information retrieval Shared task at the CLEFeHealth Lab conducted in 2014. Task 3 aimed at evaluating information retrieval to address questions patients may have when reading clinical reports. The package contains: • a collection of medical-related documents, • guidelines provided to the participants, • queries generated by medical professionals, • a set of manual relevance assessments, • the official results obtained by the participants, • working notes papers. The collection consists of a set of around 1 million medical-related documents, provided by the Khresmoi project. This collection contains documents covering a broad set of medical topics, and does not contain any patient information. The documents in the collection come from several online sources, including Health On the Net organization certified websites, as well as well-known medical sites and databases (e.g. Genetics Home Reference, Five sample development queries, 50 test queries and their translation in Czech, French and German, as well as result set are provided with the data set. The queries have been manually generated by medical professionals from disorders mentioned in discharge summaries used for task 2 of CLEF eHealth. The training set contains 20 topics: • 5 in English • 5 in Czech • 5 in French • 5 in German Topic Test Set The official test topic and result set for Task 3 consists of 50 topics in 2013 and 200 in 2014 (50 in English, 50 in Czech, 50 in French and 50 in German) and corresponding result set generated from manual relevance assessment (by medical professionals) on a pool generated from participants runs. The test set contains 200 topics: • 50 in English • 50 in Czech • 50 in French • 50 in German A demo version of the package is available from

Right Holder(s)