Abstract
Traditional Information Retrieval (IR) systems mainly focus on answering questions about events or objects. However, there are various types of question forms that require IR systems to build complex answers from multiple data sources. Therefore, the idea of building IR systems that can create complex answers automatically, became the aim of TREC CAR 2017-2019. CAR (Complex Answer Retrieval) is one of many tracks, was hosted by TREC (The Text REtrieval Conference) where is a playground for the information retrieval community.
In this paper, we built an improved complex answer retrieval system based on the system model of Nogueira et al. [3]. Our method tries to increase the coverage of the retrieval task. Thereby, the performance of our system shows that the MAP, MRR, and NDCG evaluation scores are improved.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Anserini. https://github.com/castorini/anserini.
References
Devlin, J., Chang, M.-W., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding. arXiv:1810.04805v2 (2019)
MacAvaney, S., Yates, A., Hui, K.: Contextualized PACRR for complex answer retrieval. In: TREC CAR 2017 (2017)
Nogueira, R., Cho, K.: Passage Re-ranking with BERT. arXiv:1901.04085 (2019)
Nogueira, R., Cho, K.: New York University at TREC 2018 complex answer retrieval track. In: TREC CAR 2018 (2018)
benchmarkY1-test.v2.0.tar.xz. http://trec-car.cs.unh.edu/datareleases/v2.0/benchmarkY1-test.v2.0.tar.xz. Accessed 08 May 2019
train.v2.0.tar.xz. http://trec-car.cs.unh.edu/datareleases/v2.0/train.v2.0.tar.xz. Accessed 08 May 2019
paragraphCorpus.v2.0.tar.xz. http://trec-car.cs.unh.edu/datareleases/v2.0/paragraphCorpus.v2.0.tar.xz. Accessed 08 May 2019
trec-car-tool. https://github.com/TREMA-UNH/trec-car-tools-java. Accessed 08 May 2019
trec_eval. https://github.com/usnistgov/trec_eval. Accessed 08 May 2019
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Ha, L., Nguyen, D.T. (2019). Towards an Improvement of Complex Answer Retrieval System. In: Dang, T., Küng, J., Takizawa, M., Bui, S. (eds) Future Data and Security Engineering. FDSE 2019. Lecture Notes in Computer Science(), vol 11814. Springer, Cham. https://doi.org/10.1007/978-3-030-35653-8_45
Download citation
DOI: https://doi.org/10.1007/978-3-030-35653-8_45
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-35652-1
Online ISBN: 978-3-030-35653-8
eBook Packages: Computer ScienceComputer Science (R0)