HiCAL is a system for efficient high-recall retrieval. The system allows retrieving and assessing relevant documents and provides high data processing performance and a user-friendly document assessment interface.
Our model was evaluated on the standard TREC dataset: TREC Core 2017 Track and TREC Core 2018 Track.
Visit hical.github.io for usage and installation instruction. For component specific details, check the README in their respective directory.
This repo contains the implementation of High-Recall Information Retrieval system, described in the following papers:
-
Haotian Zhang, Mustafa Abualsaud, Nimesh Ghelani, Mark Smucker, Gordon Cormack and Maura Grossman. Effective User Interaction for High-Recall Retrieval: Less is More CIKM 2018
-
Nimesh Ghelani, Gordon Cormack, and Mark Smucker. Refresh Strategies in Continuous Active Learning SIGIR 2018 workshop on Professional Search
-
Mustafa Abualsaud, Nimesh Ghelani, Haotian Zhang, Mark Smucker, Gordon Cormack and Maura Grossman. A System for Efficient High-Recall Retrieval Proceedings of the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2018)
-
Haotian Zhang, Mustafa Abualsaud, Nimesh Ghelani, Angshuman Ghosh, Mark Smucker, Gordon Cormack and Maura Grossman. UWaterlooMDS at the TREC 2017 Common Core Track (TREC 2017)
-
Haotian Zhang, Gordon Cormack, Maura Grossman and Mark Smucker. Evaluating Sentence-Level Relevance Feedback for High-Recall Information Retrieval