Focal elements of neural information retrieval models. An outlook through a reproducibility study

作者：

Highlights：

• Replicability study of a supervised neural IR model.

• Reproducibility study of an unsupervised neural IR model.

• Generalization of both neural IR models on heterogeneous and multilingual experimental collections.

• In-depth per topic analysis of the performances to individuate neural IR models strong aspects.

• Extensive evaluation on shared test collections and statistical analyses.

摘要

•Replicability study of a supervised neural IR model.•Reproducibility study of an unsupervised neural IR model.•Generalization of both neural IR models on heterogeneous and multilingual experimental collections.•In-depth per topic analysis of the performances to individuate neural IR models strong aspects.•Extensive evaluation on shared test collections and statistical analyses.

论文关键词：Neural information retrieval,Reproducibility,Neural vectors,Deep learning

论文评审过程：Received 28 February 2019, Revised 30 August 2019, Accepted 31 August 2019, Available online 13 September 2019, Version of Record 20 October 2020.

论文官网地址：https://doi.org/10.1016/j.ipm.2019.102109