Downloads

Data and utilities and other stuff.

return to home

indRi (R library)

indRi.tgz

An R library for working with a Lemur/indri index. See help("indRi") after installing.


Multiple Comparisons Adjustments for Cranfield Testing (R library)

CranfieldMCP.tgz

As described in B. Carterette, Multiple Testing in Statistical Analysis of Systems-Based Information Retrieval Experiments, ACM TOIS (to appear)


Rank Distance

drank-0.5.tar.gz (C++, requires GNU Scientific Library version >= 1.12)

As described in B. Carterette, On Rank Correlation and the Distance Between Rankings, SIGIR 2009.


Minimal Test Collection (MTC) Evaluation Utility

mtc-eval.tgz (C, requires GNU Scientific Library)

Tools for evaluation with few relevance judgments.

Used in the TREC Million Query Track; described in Minimal Test Collections for Retrieval Evaluation, Robust Test Collections for Retrieval Evaluation, Using Document Similarities for Semiautomatic Evaluation of IR Systems


Minimal Test Collection (MTC) Document Selection

mtc-server-0.5.tar.gz

Used in the TREC Million Query Track; described in Minimal Test Collections for Retrieval Evaluation