@inproceedings{lin-etal-2018-adaptive,
title = "Adaptive Scaling for Sparse Detection in Information Extraction",
author = "Lin, Hongyu and
Lu, Yaojie and
Han, Xianpei and
Sun, Le",
editor = "Gurevych, Iryna and
Miyao, Yusuke",
booktitle = "Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2018",
address = "Melbourne, Australia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/P18-1095",
doi = "10.18653/v1/P18-1095",
pages = "1033--1043",
abstract = "This paper focuses on detection tasks in information extraction, where positive instances are sparsely distributed and models are usually evaluated using F-measure on positive classes. These characteristics often result in deficient performance of neural network based detection models. In this paper, we propose \textit{adaptive scaling}, an algorithm which can handle the positive sparsity problem and directly optimize over F-measure via dynamic cost-sensitive learning. To this end, we borrow the idea of marginal utility from economics and propose a theoretical framework for instance importance measuring without introducing any additional hyper-parameters. Experiments show that our algorithm leads to a more effective and stable training of neural network based detection models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lin-etal-2018-adaptive">
<titleInfo>
<title>Adaptive Scaling for Sparse Detection in Information Extraction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hongyu</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaojie</namePart>
<namePart type="family">Lu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xianpei</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Le</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2018-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Iryna</namePart>
<namePart type="family">Gurevych</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yusuke</namePart>
<namePart type="family">Miyao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Melbourne, Australia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper focuses on detection tasks in information extraction, where positive instances are sparsely distributed and models are usually evaluated using F-measure on positive classes. These characteristics often result in deficient performance of neural network based detection models. In this paper, we propose adaptive scaling, an algorithm which can handle the positive sparsity problem and directly optimize over F-measure via dynamic cost-sensitive learning. To this end, we borrow the idea of marginal utility from economics and propose a theoretical framework for instance importance measuring without introducing any additional hyper-parameters. Experiments show that our algorithm leads to a more effective and stable training of neural network based detection models.</abstract>
<identifier type="citekey">lin-etal-2018-adaptive</identifier>
<identifier type="doi">10.18653/v1/P18-1095</identifier>
<location>
<url>https://aclanthology.org/P18-1095</url>
</location>
<part>
<date>2018-07</date>
<extent unit="page">
<start>1033</start>
<end>1043</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Adaptive Scaling for Sparse Detection in Information Extraction
%A Lin, Hongyu
%A Lu, Yaojie
%A Han, Xianpei
%A Sun, Le
%Y Gurevych, Iryna
%Y Miyao, Yusuke
%S Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2018
%8 July
%I Association for Computational Linguistics
%C Melbourne, Australia
%F lin-etal-2018-adaptive
%X This paper focuses on detection tasks in information extraction, where positive instances are sparsely distributed and models are usually evaluated using F-measure on positive classes. These characteristics often result in deficient performance of neural network based detection models. In this paper, we propose adaptive scaling, an algorithm which can handle the positive sparsity problem and directly optimize over F-measure via dynamic cost-sensitive learning. To this end, we borrow the idea of marginal utility from economics and propose a theoretical framework for instance importance measuring without introducing any additional hyper-parameters. Experiments show that our algorithm leads to a more effective and stable training of neural network based detection models.
%R 10.18653/v1/P18-1095
%U https://aclanthology.org/P18-1095
%U https://doi.org/10.18653/v1/P18-1095
%P 1033-1043
Markdown (Informal)
[Adaptive Scaling for Sparse Detection in Information Extraction](https://aclanthology.org/P18-1095) (Lin et al., ACL 2018)
ACL
- Hongyu Lin, Yaojie Lu, Xianpei Han, and Le Sun. 2018. Adaptive Scaling for Sparse Detection in Information Extraction. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1033–1043, Melbourne, Australia. Association for Computational Linguistics.