@inproceedings{chalkidis-etal-2020-empirical,
title = "An Empirical Study on Large-Scale Multi-Label Text Classification Including Few and Zero-Shot Labels",
author = "Chalkidis, Ilias and
Fergadiotis, Manos and
Kotitsas, Sotiris and
Malakasiotis, Prodromos and
Aletras, Nikolaos and
Androutsopoulos, Ion",
editor = "Webber, Bonnie and
Cohn, Trevor and
He, Yulan and
Liu, Yang",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.emnlp-main.607",
doi = "10.18653/v1/2020.emnlp-main.607",
pages = "7503--7515",
abstract = "Large-scale Multi-label Text Classification (LMTC) has a wide range of Natural Language Processing (NLP) applications and presents interesting challenges. First, not all labels are well represented in the training set, due to the very large label set and the skewed label distributions of datasets. Also, label hierarchies and differences in human labelling guidelines may affect graph-aware annotation proximity. Finally, the label hierarchies are periodically updated, requiring LMTC models capable of zero-shot generalization. Current state-of-the-art LMTC models employ Label-Wise Attention Networks (LWANs), which (1) typically treat LMTC as flat multi-label classification; (2) may use the label hierarchy to improve zero-shot learning, although this practice is vastly understudied; and (3) have not been combined with pre-trained Transformers (e.g. BERT), which have led to state-of-the-art results in several NLP benchmarks. Here, for the first time, we empirically evaluate a battery of LMTC methods from vanilla LWANs to hierarchical classification approaches and transfer learning, on frequent, few, and zero-shot learning on three datasets from different domains. We show that hierarchical methods based on Probabilistic Label Trees (PLTs) outperform LWANs. Furthermore, we show that Transformer-based approaches outperform the state-of-the-art in two of the datasets, and we propose a new state-of-the-art method which combines BERT with LWAN. Finally, we propose new models that leverage the label hierarchy to improve few and zero-shot learning, considering on each dataset a graph-aware annotation proximity measure that we introduce.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chalkidis-etal-2020-empirical">
<titleInfo>
<title>An Empirical Study on Large-Scale Multi-Label Text Classification Including Few and Zero-Shot Labels</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ilias</namePart>
<namePart type="family">Chalkidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manos</namePart>
<namePart type="family">Fergadiotis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sotiris</namePart>
<namePart type="family">Kotitsas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Prodromos</namePart>
<namePart type="family">Malakasiotis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolaos</namePart>
<namePart type="family">Aletras</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ion</namePart>
<namePart type="family">Androutsopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bonnie</namePart>
<namePart type="family">Webber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Trevor</namePart>
<namePart type="family">Cohn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yulan</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large-scale Multi-label Text Classification (LMTC) has a wide range of Natural Language Processing (NLP) applications and presents interesting challenges. First, not all labels are well represented in the training set, due to the very large label set and the skewed label distributions of datasets. Also, label hierarchies and differences in human labelling guidelines may affect graph-aware annotation proximity. Finally, the label hierarchies are periodically updated, requiring LMTC models capable of zero-shot generalization. Current state-of-the-art LMTC models employ Label-Wise Attention Networks (LWANs), which (1) typically treat LMTC as flat multi-label classification; (2) may use the label hierarchy to improve zero-shot learning, although this practice is vastly understudied; and (3) have not been combined with pre-trained Transformers (e.g. BERT), which have led to state-of-the-art results in several NLP benchmarks. Here, for the first time, we empirically evaluate a battery of LMTC methods from vanilla LWANs to hierarchical classification approaches and transfer learning, on frequent, few, and zero-shot learning on three datasets from different domains. We show that hierarchical methods based on Probabilistic Label Trees (PLTs) outperform LWANs. Furthermore, we show that Transformer-based approaches outperform the state-of-the-art in two of the datasets, and we propose a new state-of-the-art method which combines BERT with LWAN. Finally, we propose new models that leverage the label hierarchy to improve few and zero-shot learning, considering on each dataset a graph-aware annotation proximity measure that we introduce.</abstract>
<identifier type="citekey">chalkidis-etal-2020-empirical</identifier>
<identifier type="doi">10.18653/v1/2020.emnlp-main.607</identifier>
<location>
<url>https://aclanthology.org/2020.emnlp-main.607</url>
</location>
<part>
<date>2020-11</date>
<extent unit="page">
<start>7503</start>
<end>7515</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T An Empirical Study on Large-Scale Multi-Label Text Classification Including Few and Zero-Shot Labels
%A Chalkidis, Ilias
%A Fergadiotis, Manos
%A Kotitsas, Sotiris
%A Malakasiotis, Prodromos
%A Aletras, Nikolaos
%A Androutsopoulos, Ion
%Y Webber, Bonnie
%Y Cohn, Trevor
%Y He, Yulan
%Y Liu, Yang
%S Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
%D 2020
%8 November
%I Association for Computational Linguistics
%C Online
%F chalkidis-etal-2020-empirical
%X Large-scale Multi-label Text Classification (LMTC) has a wide range of Natural Language Processing (NLP) applications and presents interesting challenges. First, not all labels are well represented in the training set, due to the very large label set and the skewed label distributions of datasets. Also, label hierarchies and differences in human labelling guidelines may affect graph-aware annotation proximity. Finally, the label hierarchies are periodically updated, requiring LMTC models capable of zero-shot generalization. Current state-of-the-art LMTC models employ Label-Wise Attention Networks (LWANs), which (1) typically treat LMTC as flat multi-label classification; (2) may use the label hierarchy to improve zero-shot learning, although this practice is vastly understudied; and (3) have not been combined with pre-trained Transformers (e.g. BERT), which have led to state-of-the-art results in several NLP benchmarks. Here, for the first time, we empirically evaluate a battery of LMTC methods from vanilla LWANs to hierarchical classification approaches and transfer learning, on frequent, few, and zero-shot learning on three datasets from different domains. We show that hierarchical methods based on Probabilistic Label Trees (PLTs) outperform LWANs. Furthermore, we show that Transformer-based approaches outperform the state-of-the-art in two of the datasets, and we propose a new state-of-the-art method which combines BERT with LWAN. Finally, we propose new models that leverage the label hierarchy to improve few and zero-shot learning, considering on each dataset a graph-aware annotation proximity measure that we introduce.
%R 10.18653/v1/2020.emnlp-main.607
%U https://aclanthology.org/2020.emnlp-main.607
%U https://doi.org/10.18653/v1/2020.emnlp-main.607
%P 7503-7515
Markdown (Informal)
[An Empirical Study on Large-Scale Multi-Label Text Classification Including Few and Zero-Shot Labels](https://aclanthology.org/2020.emnlp-main.607) (Chalkidis et al., EMNLP 2020)
ACL