High Accuracy Retrieval from Documents (HARD) 2004

The objective of the HARD 2004 project is to achieve high accuracy retrieval from documents by leveraging additional information about the searcher and/or the search context, through techniques such as passage retrieval, and using very targeted interaction with the searcher. HARD is in its second year, and is being run as an evaluation track within TREC, the Text REtrieval Conference sponsored by NIST.

The manner in which HARD researchers will leverage additional information about the searcher is through the use of "Metadata", which in this case means that potential searchers will select an array of specific values that will further refine their initial queries. LDC annotators will be responsible for topic creation, and will use the Topic Creation Web Interface to do so.

Project Links

2004 Project Timeline
Visit this page for more information about progress and deliverables.

Intranet HARD page
Annotators: Visit this page for more information about creating and submitting topics

Clarification Form Submission Instructions
Instructions on submitting clarification forms

2004 Guidelines
As of October 11, 2004: Guidelines for 2004 are available in .pdf format.

HARD 2003
The HARD track was a pilot track in 2003; for more information about the first year of HARD annotation, please visit the HARD 2003 webpage.

2003 Guidelines
HARD 2003 annotation guidelines available in pdf format.

HARD Project Overview
HARD project overview and guidelines for researchers at the University of Massachussetts

NIST's website

LDC's TIDES page
This site provides an overview of all TIDES projects that LDC supports, in addition to a link to the TIDES data matrix.