Multi-Class Detection of Abusive Language Using Automated Machine Learning

Mackenzie Jorgensen; Minho Choi; Marco Niemann; Jens Brunk; Jörg Becker

doi:10.30844/wi_2020_r7-jorgensen

Multi-Class Detection of Abusive Language Using Automated Machine Learning

Mackenzie Jorgensen, Minho Choi, Marco Niemann, Jens Brunk, Jörg Becker

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Abusive language detection online is a daunting task for moderators. We propose Automated Machine Learning (Auto-ML) to semi-automate abusive language detection and to assist moderators. In this paper, we show that multi-class classification powered by Auto-ML is successful in detecting abusive language in English and German as well as and better than the state-ofthe- art machine learning models. We also highlight how we combatted the imbalanced data problem in our data-sets through feature selection and undersampling methods. We propose Auto-ML as a promising approach to the field of abusive language detection, especially for small companies who may have little machine learning knowledge and computing resources.

Original language	English
Title of host publication	WI2020 Zentrale Tracks
Subtitle of host publication	Changing landscapes
Place of Publication	Potstdam, Germany
Publisher	International Association for Safe & Ethical AI
Pages	1763-1775
DOIs	https://doi.org/10.30844/wi_2020_r7-jorgensen
Publication status	Published - 9 Mar 2020
Externally published	Yes
Event	15th International Conference on Wirtschaftsinformatik - Potsdam, Germany Duration: 8 Mar 2020 → 11 Mar 2020

Conference

Conference	15th International Conference on Wirtschaftsinformatik
Country/Territory	Germany
City	Potsdam
Period	8/03/20 → 11/03/20

Keywords

Abusive Language Detection
Multi-Class Classification
Automated-Machine Learning

Access to Document

10.30844/wi_2020_r7-jorgensen

Jorgensenetal-Multi-ClassDetectionofAbusiveLanguageUsingAutoML (1)Final published version, 770 KB

Cite this

@inproceedings{97775700c4a04390a274ea0a9916261e,

title = "Multi-Class Detection of Abusive Language Using Automated Machine Learning",

abstract = "Abusive language detection online is a daunting task for moderators. We propose Automated Machine Learning (Auto-ML) to semi-automate abusive language detection and to assist moderators. In this paper, we show that multi-class classification powered by Auto-ML is successful in detecting abusive language in English and German as well as and better than the state-ofthe- art machine learning models. We also highlight how we combatted the imbalanced data problem in our data-sets through feature selection and undersampling methods. We propose Auto-ML as a promising approach to the field of abusive language detection, especially for small companies who may have little machine learning knowledge and computing resources.",

keywords = "Abusive Language Detection, Multi-Class Classification, Automated-Machine Learning",

author = "Mackenzie Jorgensen and Minho Choi and Marco Niemann and Jens Brunk and J{\"o}rg Becker",

year = "2020",

month = mar,

day = "9",

doi = "10.30844/wi\_2020\_r7-jorgensen",

language = "English",

pages = "1763--1775",

booktitle = "WI2020 Zentrale Tracks",

publisher = "International Association for Safe \& Ethical AI ",

note = "15th International Conference on Wirtschaftsinformatik ; Conference date: 08-03-2020 Through 11-03-2020",

}

Jorgensen, M, Choi, M, Niemann, M, Brunk, J & Becker, J 2020, Multi-Class Detection of Abusive Language Using Automated Machine Learning. in WI2020 Zentrale Tracks: Changing landscapes. International Association for Safe & Ethical AI , Potstdam, Germany, pp. 1763-1775, 15th International Conference on Wirtschaftsinformatik, Potsdam, Germany, 8/03/20. https://doi.org/10.30844/wi_2020_r7-jorgensen

Multi-Class Detection of Abusive Language Using Automated Machine Learning. / Jorgensen, Mackenzie; Choi, Minho; Niemann, Marco et al.
WI2020 Zentrale Tracks: Changing landscapes. Potstdam, Germany: International Association for Safe & Ethical AI , 2020. p. 1763-1775.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Multi-Class Detection of Abusive Language Using Automated Machine Learning

AU - Jorgensen, Mackenzie

AU - Choi, Minho

AU - Niemann, Marco

AU - Brunk, Jens

AU - Becker, Jörg

PY - 2020/3/9

Y1 - 2020/3/9

N2 - Abusive language detection online is a daunting task for moderators. We propose Automated Machine Learning (Auto-ML) to semi-automate abusive language detection and to assist moderators. In this paper, we show that multi-class classification powered by Auto-ML is successful in detecting abusive language in English and German as well as and better than the state-ofthe- art machine learning models. We also highlight how we combatted the imbalanced data problem in our data-sets through feature selection and undersampling methods. We propose Auto-ML as a promising approach to the field of abusive language detection, especially for small companies who may have little machine learning knowledge and computing resources.

AB - Abusive language detection online is a daunting task for moderators. We propose Automated Machine Learning (Auto-ML) to semi-automate abusive language detection and to assist moderators. In this paper, we show that multi-class classification powered by Auto-ML is successful in detecting abusive language in English and German as well as and better than the state-ofthe- art machine learning models. We also highlight how we combatted the imbalanced data problem in our data-sets through feature selection and undersampling methods. We propose Auto-ML as a promising approach to the field of abusive language detection, especially for small companies who may have little machine learning knowledge and computing resources.

KW - Abusive Language Detection

KW - Multi-Class Classification

KW - Automated-Machine Learning

U2 - 10.30844/wi_2020_r7-jorgensen

DO - 10.30844/wi_2020_r7-jorgensen

M3 - Conference contribution

SP - 1763

EP - 1775

BT - WI2020 Zentrale Tracks

PB - International Association for Safe & Ethical AI

CY - Potstdam, Germany

T2 - 15th International Conference on Wirtschaftsinformatik

Y2 - 8 March 2020 through 11 March 2020

ER -

Multi-Class Detection of Abusive Language Using Automated Machine Learning

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this