<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="serra-etal-2017-class">
    <titleInfo>
        <title>Class-based Prediction Errors to Detect Hate Speech with Out-of-vocabulary Words</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Joan</namePart>
        <namePart type="family">Serrà</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ilias</namePart>
        <namePart type="family">Leontiadis</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dimitris</namePart>
        <namePart type="family">Spathis</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Gianluca</namePart>
        <namePart type="family">Stringhini</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jeremy</namePart>
        <namePart type="family">Blackburn</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Athena</namePart>
        <namePart type="family">Vakali</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2017-08</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the First Workshop on Abusive Language Online</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Zeerak</namePart>
            <namePart type="family">Waseem</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Wendy</namePart>
            <namePart type="given">Hui</namePart>
            <namePart type="given">Kyong</namePart>
            <namePart type="family">Chung</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Dirk</namePart>
            <namePart type="family">Hovy</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Joel</namePart>
            <namePart type="family">Tetreault</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Vancouver, BC, Canada</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Common approaches to text categorization essentially rely either on n-gram counts or on word embeddings. This presents important difficulties in highly dynamic or quickly-interacting environments, where the appearance of new words and/or varied misspellings is the norm. A paradigmatic example of this situation is abusive online behavior, with social networks and media platforms struggling to effectively combat uncommon or non-blacklisted hate words. To better deal with these issues in those fast-paced environments, we propose using the error signal of class-based language models as input to text classification algorithms. In particular, we train a next-character prediction model for any given class and then exploit the error of such class-based models to inform a neural network classifier. This way, we shift from the ‘ability to describe’ seen documents to the ‘ability to predict’ unseen content. Preliminary studies using out-of-vocabulary splits from abusive tweet data show promising results, outperforming competitive text categorization strategies by 4-11%.</abstract>
    <identifier type="citekey">serra-etal-2017-class</identifier>
    <identifier type="doi">10.18653/v1/W17-3005</identifier>
    <location>
        <url>https://aclanthology.org/W17-3005</url>
    </location>
    <part>
        <date>2017-08</date>
        <extent unit="page">
            <start>36</start>
            <end>40</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Class-based Prediction Errors to Detect Hate Speech with Out-of-vocabulary Words
%A Serrà, Joan
%A Leontiadis, Ilias
%A Spathis, Dimitris
%A Stringhini, Gianluca
%A Blackburn, Jeremy
%A Vakali, Athena
%Y Waseem, Zeerak
%Y Chung, Wendy Hui Kyong
%Y Hovy, Dirk
%Y Tetreault, Joel
%S Proceedings of the First Workshop on Abusive Language Online
%D 2017
%8 August
%I Association for Computational Linguistics
%C Vancouver, BC, Canada
%F serra-etal-2017-class
%X Common approaches to text categorization essentially rely either on n-gram counts or on word embeddings. This presents important difficulties in highly dynamic or quickly-interacting environments, where the appearance of new words and/or varied misspellings is the norm. A paradigmatic example of this situation is abusive online behavior, with social networks and media platforms struggling to effectively combat uncommon or non-blacklisted hate words. To better deal with these issues in those fast-paced environments, we propose using the error signal of class-based language models as input to text classification algorithms. In particular, we train a next-character prediction model for any given class and then exploit the error of such class-based models to inform a neural network classifier. This way, we shift from the ‘ability to describe’ seen documents to the ‘ability to predict’ unseen content. Preliminary studies using out-of-vocabulary splits from abusive tweet data show promising results, outperforming competitive text categorization strategies by 4-11%.
%R 10.18653/v1/W17-3005
%U https://aclanthology.org/W17-3005
%U https://doi.org/10.18653/v1/W17-3005
%P 36-40

Markdown (Informal)

[Class-based Prediction Errors to Detect Hate Speech with Out-of-vocabulary Words](https://aclanthology.org/W17-3005) (Serrà et al., ALW 2017)

• Class-based Prediction Errors to Detect Hate Speech with Out-of-vocabulary Words (Serrà et al., ALW 2017)

ACL

• Joan Serrà, Ilias Leontiadis, Dimitris Spathis, Gianluca Stringhini, Jeremy Blackburn, and Athena Vakali. 2017. Class-based Prediction Errors to Detect Hate Speech with Out-of-vocabulary Words. In Proceedings of the First Workshop on Abusive Language Online, pages 36–40, Vancouver, BC, Canada. Association for Computational Linguistics.