CLEF QAST (2007-2009) – Evaluation Package

Full Official Name: CLEF QAST (2007-2009) – Evaluation Package
Submission date: Jan. 24, 2014, 4:22 p.m.

The Cross-Language Evaluation Forum (CLEF) promotes R&D in multilingual information access (MLIA) by (i) developing an infrastructure for the testing, tuning and evaluation of information retrieval systems operating on European languages in both monolingual and cross-language contexts, and (ii) creating test-suites of reusable data which can be employed by system developers for benchmarking purposes. CLEF QAST (2007-2009) contains the data used for the Question Answering on Speech Transcripts tracks of the CLEF campaigns carried out from 2007 to 2009. These tracks tested the performance of monolingual Question Answering systems on collections of audio transcriptions. The CLEF Test Suite is composed of: • Transcript Data Collections • Questions • Guidelines • Relevance assessments • Official campaign results • Working notes papers The Transcript Data Collections consist of manual transcriptions and ASR automatic transcriptions from the following development and evaluation corpora: • English datasets: o CHIL Collection (2007-2008 tracks) (see also ELRA-E0009, E0010, E0017 and E0033 for CHIL Evaluation Packages) o AMI Collection (2007-2008 tracks) o European Parliament from 2005 TC-STAR Evaluation campaign (2008-2009 tracks) (see also ELRA-E0002) • Spanish datasets: European Parliament from 2005 TC-STAR Evaluation campaign (2008-2009 tracks) (see also ELRA-E0003) • French dataset: ESTER Corpus (2008-2009 tracks) (see also ELRA-S0241) The full package is stored on 1 CD.

Creator(s)
Distributor(s)
Right Holder(s)