Multi-label Classification for Hate Speech and Abusive Language in Indonesian-Local Languages

被引:2
作者
Asti, Ajeng Dwi [1 ]
Budi, Indra [1 ]
Ibrohim, Muhammad Okky [1 ]
机构
[1] Univ Indonesia, Fac Comp Sci, Depok, Indonesia
来源
13TH INTERNATIONAL CONFERENCE ON ADVANCED COMPUTER SCIENCE AND INFORMATION SYSTEMS (ICACSIS 2021) | 2021年
关键词
hate speech; multi-label classification; Indonesian local language; Twitter;
D O I
10.1109/ICACSIS53237.2021.9631316
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Each hate speech has a target, category, and level that needs to be detected to help the authorities prioritize hate speech cases that need to be solved first. Various studies have been conducted in Indonesia on abusive speech and hate speech and their targets, categories, and levels, but only in Indonesian and English. On the other hand, various local languages in Indonesia open up opportunities for hate speech to occur using the local language. This study aims to compare some of the best machine learning algorithms, transformation methods, and feature extraction techniques in classifying abusive language and hate speech and their targets, categories, and levels using Twitter data in Indonesian and local languages. This study uses five local languages in Indonesia with the most speakers: Javanese, Sundanese, Madurese, Minangkabau, and Musi (Palembang). The algorithms used are Support Vector Machine (SVM), Multinomial Naive Bayes (MNB), and Random Forest Decision Tree (RFDT) with Binary Relevance (BR), Classifier Chains (CC), and Label Powerset (LP) as transformation methods. The term weighting used in this study is TF-IDF with word n-gram and char n-gram features. The results showed that the SVM algorithm with the CC transformation method and unigram feature extraction gave the highest F1-score results, 66.33% for Javanese and 65.68% for Sundanese. In Madurese, Minangkabau, and Musi language data, the best F1-score was obtained using the RFDT algorithm with the CC transformation method and unigram feature extraction with F1-score 76.37% 80.75%, and 77.34%.
引用
收藏
页码:325 / 330
页数:6
相关论文
共 50 条
[31]   Multi-label Classification on Natural Language Sentences for Video Game Design [J].
Zhan, Yue ;
Hsiao, Michael S. .
2019 IEEE INTERNATIONAL CONFERENCE ON HUMANIZED COMPUTING AND COMMUNICATION (HCC 2019), 2019, :52-59
[32]   OPTIMAL RANKING IN MULTI-LABEL CLASSIFICATION USING LOCAL PRECISION RATES [J].
Jiang, Ci-Ren ;
Liu, Chun-Chi ;
Zhou, Xianghong J. ;
Huang, Haiyan .
STATISTICA SINICA, 2014, 24 (04) :1547-1570
[33]   Local analgesia adverse effects prediction using multi-label classification [J].
Qu, Guangzhi ;
Wu, Hui ;
Hartrick, Craig T. ;
Niu, Jianwei .
NEUROCOMPUTING, 2012, 92 :18-27
[34]   How well do hate speech, toxicity, abusive and offensive language classification models generalize across datasets? [J].
Fortuna, Paula ;
Soler-Company, Juan ;
Wanner, Leo .
INFORMATION PROCESSING & MANAGEMENT, 2021, 58 (03)
[35]   Multi-dimensional multi-label classification: Towards encompassing heterogeneous label spaces and multi-label annotations [J].
Jia, Bin -Bin ;
Zhang, Min -Ling .
PATTERN RECOGNITION, 2023, 138
[36]   Multi-label classification with a reject option [J].
Pillai, Ignazio ;
Fumera, Giorgio ;
Roli, Fabio .
PATTERN RECOGNITION, 2013, 46 (08) :2256-2266
[37]   Compact learning for multi-label classification [J].
Lv, Jiaqi ;
Wu, Tianran ;
Peng, Chenglun ;
Liu, Yunpeng ;
Xu, Ning ;
Geng, Xin .
PATTERN RECOGNITION, 2021, 113
[38]   Interdependence Model for Multi-label Classification [J].
Yoshimura, Kosuke ;
Iwase, Tomoaki ;
Baba, Yukino ;
Kashima, Hisashi .
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2019: TEXT AND TIME SERIES, PT IV, 2019, 11730 :55-68
[39]   Classifier chains for multi-label classification [J].
Jesse Read ;
Bernhard Pfahringer ;
Geoff Holmes ;
Eibe Frank .
Machine Learning, 2011, 85
[40]   Source Detection With Multi-Label Classification [J].
Vijayamohanan, Jayakrishnan ;
Gupta, Arjun ;
Noakoasteen, Oameed ;
Goudos, Sotirios K. K. ;
Christodoulou, Christos G. .
IEEE OPEN JOURNAL OF SIGNAL PROCESSING, 2023, 4 :336-345