amourao / federator

Rank fusion library that takes ranks in the TREC format and can combine them using multiple types of score and rank based rank fusion techniques.

Federator

Rank fusion library that takes ranks in the TREC format and can combine them using multiple types of score and rank based rank fusion techniques (e.g. CombSUM and variants, RRF, ISR and variants).

It can also be easily extented to allow for novel rank fusion techniques (check the examples in the fusion_functions.py).

This library was used as a part of multiple research articles:

A. Mourão, J. Magalhães, “Low-Complexity Supervised Rank Fusion Models”. CIKM 2018.

A. Mourão, F. Martins, J. Magalhães, “NovaSearch at TREC 2015 Clinical Decision Support Track”, TREC (Working Notes). 2015.

A. Mourão, F. Martins, J. Magalhães, “Multimodal medical information retrieval with unsupervised rank fusion”, Computerized Medical Imaging and Graphics 2015.

A. Mourão, J. Magalhães, “Scalable multimodal search with distributed indexing by sparse hashing”. ICMR 2015.

A. Mourão, F. Martins, J. Magalhães, “NovaSearch at TREC 2014 Clinical Decision Support Track”, TREC (Working Notes). 2014.

A. Mourão, F. Martins, J. Magalhães, “NovaSearch at TREC 2013 Federated Web Search Track: Experiments with rank fusion”, TREC (Working Notes). 2013.

A. Mourão, F. Martins, J. Magalhães, “NovaSearch on Medical ImageCLEF 2013”, CLEF (Working Notes). 2013.