@inproceedings{nejadgholi-etal-2023-concept,
title = "Concept-Based Explanations to Test for False Causal Relationships Learned by Abusive Language Classifiers",
author = "Nejadgholi, Isar and
Kiritchenko, Svetlana and
Fraser, Kathleen C. and
Balkir, Esma",
editor = "Chung, Yi-ling and
R{\{}{\textbackslash}{''}ottger{\}}, Paul and
Nozza, Debora and
Talat, Zeerak and
Mostafazadeh Davani, Aida",
booktitle = "The 7th Workshop on Online Abuse and Harms (WOAH)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.woah-1.14/",
doi = "10.18653/v1/2023.woah-1.14",
pages = "138--149",
abstract = "Classifiers tend to learn a false causal relationship between an over-represented concept and a label, which can result in over-reliance on the concept and compromised classification accuracy. It is imperative to have methods in place that can compare different models and identify over-reliances on specific concepts. We consider three well-known abusive language classifiers trained on large English datasets and focus on the concept of negative emotions, which is an important signal but should not be learned as a sufficient feature for the label of abuse. Motivated by the definition of global sufficiency, we first examine the unwanted dependencies learned by the classifiers by assessing their accuracy on a challenge set across all decision thresholds. Further, recognizing that a challenge set might not always be available, we introduce concept-based explanation metrics to assess the influence of the concept on the labels. These explanations allow us to compare classifiers regarding the degree of false global sufficiency they have learned between a concept and a label."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="nejadgholi-etal-2023-concept">
<titleInfo>
<title>Concept-Based Explanations to Test for False Causal Relationships Learned by Abusive Language Classifiers</title>
</titleInfo>
<name type="personal">
<namePart type="given">Isar</namePart>
<namePart type="family">Nejadgholi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Svetlana</namePart>
<namePart type="family">Kiritchenko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kathleen</namePart>
<namePart type="given">C</namePart>
<namePart type="family">Fraser</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Esma</namePart>
<namePart type="family">Balkir</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>The 7th Workshop on Online Abuse and Harms (WOAH)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yi-ling</namePart>
<namePart type="family">Chung</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="family">R{\textbackslash”ottger}</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debora</namePart>
<namePart type="family">Nozza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zeerak</namePart>
<namePart type="family">Talat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aida</namePart>
<namePart type="family">Mostafazadeh Davani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Classifiers tend to learn a false causal relationship between an over-represented concept and a label, which can result in over-reliance on the concept and compromised classification accuracy. It is imperative to have methods in place that can compare different models and identify over-reliances on specific concepts. We consider three well-known abusive language classifiers trained on large English datasets and focus on the concept of negative emotions, which is an important signal but should not be learned as a sufficient feature for the label of abuse. Motivated by the definition of global sufficiency, we first examine the unwanted dependencies learned by the classifiers by assessing their accuracy on a challenge set across all decision thresholds. Further, recognizing that a challenge set might not always be available, we introduce concept-based explanation metrics to assess the influence of the concept on the labels. These explanations allow us to compare classifiers regarding the degree of false global sufficiency they have learned between a concept and a label.</abstract>
<identifier type="citekey">nejadgholi-etal-2023-concept</identifier>
<identifier type="doi">10.18653/v1/2023.woah-1.14</identifier>
<location>
<url>https://aclanthology.org/2023.woah-1.14/</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>138</start>
<end>149</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Concept-Based Explanations to Test for False Causal Relationships Learned by Abusive Language Classifiers
%A Nejadgholi, Isar
%A Kiritchenko, Svetlana
%A Fraser, Kathleen C.
%A Balkir, Esma
%Y Chung, Yi-ling
%Y R{\textbackslash”ottger}, Paul
%Y Nozza, Debora
%Y Talat, Zeerak
%Y Mostafazadeh Davani, Aida
%S The 7th Workshop on Online Abuse and Harms (WOAH)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F nejadgholi-etal-2023-concept
%X Classifiers tend to learn a false causal relationship between an over-represented concept and a label, which can result in over-reliance on the concept and compromised classification accuracy. It is imperative to have methods in place that can compare different models and identify over-reliances on specific concepts. We consider three well-known abusive language classifiers trained on large English datasets and focus on the concept of negative emotions, which is an important signal but should not be learned as a sufficient feature for the label of abuse. Motivated by the definition of global sufficiency, we first examine the unwanted dependencies learned by the classifiers by assessing their accuracy on a challenge set across all decision thresholds. Further, recognizing that a challenge set might not always be available, we introduce concept-based explanation metrics to assess the influence of the concept on the labels. These explanations allow us to compare classifiers regarding the degree of false global sufficiency they have learned between a concept and a label.
%R 10.18653/v1/2023.woah-1.14
%U https://aclanthology.org/2023.woah-1.14/
%U https://doi.org/10.18653/v1/2023.woah-1.14
%P 138-149
Markdown (Informal)
[Concept-Based Explanations to Test for False Causal Relationships Learned by Abusive Language Classifiers](https://aclanthology.org/2023.woah-1.14/) (Nejadgholi et al., WOAH 2023)
ACL