A Lexicon-based Approach for Hate Speech Detection

Gitari, Njagi Dennis; Zhang, Zuping; Hanyurwimfura, Damien; Long, Jun

doi:10.14257/ijmue.2015.10.4.21

Cited by 353 publications

(169 citation statements)

References 21 publications

Supporting

Mentioning

161

Contrasting

Unclassified

Order By: Relevance

“…These instruments measure the sentiment of words (e.g., angry, sad, excellent, depressed) in order to calculate sentiment of document (Pak & Paroubek, n.d.; O'Connor, Balasubramanyan, Routledge, & Smith, ; Saif et al, ; Saif, Fernandez, He, & Alani, ). In fact, such dictionary contains words tagged with both a polarity label and semantic orientation score or reliability label (Gitari, Zuping, Damien, & Long, ; Taboada, Anthony, & Voll, ; Wilson, Wiebe, & Hoffmann, ).…”

Section: Sentiment Analysis Methodsmentioning

confidence: 99%

Sentiment analysis for mining texts and social networks data: Methods and tools

Zucco

Calabrese

Agapito

et al. 2019

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Social networks (SNs) represent an established environment in which users share daily emotions and opinions. Therefore, they have become an essential source of big data related to sentiment/opinion sphere. Sentiment analysis (SA) aims to extract sentiments, emotions or opinions from texts, made available by different data sources like SNs. This review presents a depth study relative to the methods and the main tools for SA. The analysis was performed by defining four criteria and several variables to compare 24 tools with objective criteria. Specifically, the tools have been analyzed and tested to verify their usability, flexibility of use, and other specifications related to the type of analysis performed. The majority of tools can detect positive, negative, and neutral polarity, while few tools only detect positive and negative polarity. Moreover, seven tools were able to recognize emotions, and only one provides a visual map for geo‐referenced data. Except for one, remaining 23 tools offer service through the web interface. Finally, only nine tools provide both application program interfaces and a client for common programming languages to allow potential developer end‐users to integrate a specific SA tool into their application. Differently, from other recent surveys, the paper presents and discusses both methods and tools for analyzing texts and SN data sources to extract sentiment. Moreover, it contains a comprehensive comparison with other recent surveys. The comparative analysis of the tools completed according to objective criteria allows to highlight some limits on main tools that need to be faced with enhancing the end‐user experience. This article is categorized under: Technologies > Structure Discovery and Clustering Algorithmic Development > Text Mining Fundamental Concepts of Data and Knowledge > Big Data Mining Application Areas > Data Mining Software Tools

show abstract

Section: Sentiment Analysis Methodsmentioning

confidence: 99%

Sentiment analysis for mining texts and social networks data: Methods and tools

Zucco

Calabrese

Agapito

et al. 2019

WIREs Data Min & Knowl

View full text Add to dashboard Cite

show abstract

“…Nela as técnicas são aplicadas para identificar textos positivos (sem preconceito, normalmente chamado de texto neutro) ou negativo (texto com comentários preconceituosos. Existem dicionários léxicos específicos para termos preconceituosos [Gitari et al 2015, Tulkens et al 2015.…”

Section: Mineração De Opinião E Identificação De Preconceitounclassified

“…Outros trabalhos utilizam dicionários léxicos para realizar a análise, ambos tratam principalmente preconceito relacionados a raça, nacionalidade e religião [Gitari et al 2015, Tulkens et al 2015. Gitari et al propõe a extensão de dicionários léxicos como WordNet e SentiWordNet para a criação de um novo focado na identicfãoica de palavras relacionadas a preconceito.…”

Section: Trabalhos Relacionadosunclassified

See 1 more Smart Citation

Uma abordagem computacional de análise de opinião para identificação de preconceito em redações

Neto

Cavalcanti

Costa³

et al. 2017

Anais Do XXVIII Simpósio Brasileiro De Informática Na Educação (SBIE 2017)

View full text Add to dashboard Cite

Abstract. The large adoption of the ENEM grade as an instrument to evaluate students in higher education, the number of people performing this test increased significantly. Among the evaluation methods, this test includes an essay. In this exam is required to respect human rights. In this context, this work proposes the analysis of different algorithms for the identification of hate speech in essays. It adopts a hybrid approach combining machine learning techniques and lexical dictionaries. Several experiments were conducted and in the best case the proposal reached a f-measure of 82.10%. Finally, a qualitative study was carried out where essays evaluators indicated that the adoption of the proposed tool would aid in the correction of essays.Resumo. Com a grande adoção da nota ENEM como instrumento para avaliar estudantes no ingresso do ensino superior, o número de pessoas que passou a fazer essa prova aumentou significativamente. Dentre as etapas da prova existe uma redação que avalia, entre outras coisas, se o estudante fez o seu texto respeitando os direitos humanos. Diante disto, este trabalho propõe a análise de diferentes algoritmos para identificação de frases preconceituosas em redações. Para isso, foi utilizada uma abordagem híbrida combinando técnicas de aprendizagem de máquina e dicionários léxicos. Foram realizados diversos experimentos, no melhor caso a proposta atingiu uma f-measure de 82,10%. Por fim, foi realizado um estudo qualitativo onde avaliadores de redação indicaram que a adoção da ferramenta proposta auxiliaria na correção de redações.

show abstract

“…In all, there seems to be a pattern shared by most of the literature consulted (Nockleby, 2000;Djuric et al, 2015;Gitari et al, 2015;Nobata et al, 2016;Silva et al, 2016;Davidson et al, 2017), which would define hate speech as a) a deliberate attack, b) directed towards a specific group of people, and c) motivated by actual or perceived aspects that form the group's identity.…”

Section: Introductionmentioning

confidence: 99%

Hate Speech Dataset from a White Supremacy Forum

Gibert¹,

Pérez²,

García-Pablos³

et al. 2018

Proceedings of the 2nd Workshop on Abusive Language Online (ALW2)

282

166

View full text Add to dashboard Cite

Hate speech is commonly defined as any communication that disparages a target group of people based on some characteristic such as race, colour, ethnicity, gender, sexual orientation, nationality, religion, or other characteristic. Due to the massive rise of user-generated web content on social media, the amount of hate speech is also steadily increasing. Over the past years, interest in online hate speech detection and, particularly, the automation of this task has continuously grown, along with the societal impact of the phenomenon. This paper describes a hate speech dataset composed of thousands of sentences manually labelled as containing hate speech or not. The sentences have been extracted from Stormfront, a white supremacist forum. A custom annotation tool has been developed to carry out the manual labelling task which, among other things, allows the annotators to choose whether to read the context of a sentence before labelling it. The paper also provides a thoughtful qualitative and quantitative study of the resulting dataset and several baseline experiments with different classification models. The dataset is publicly available. 2 The examples in this work may contain offensive language. They have been taken from actual web data and by no means reflect the authors' opinion.3 https://github.com/aitor-garcia-p/hate-speech-dataset 4 www.stormfront.org

show abstract

A Lexicon-based Approach for Hate Speech Detection

Cited by 353 publications

References 21 publications

Sentiment analysis for mining texts and social networks data: Methods and tools

Sentiment analysis for mining texts and social networks data: Methods and tools

Uma abordagem computacional de análise de opinião para identificação de preconceito em redações

Hate Speech Dataset from a White Supremacy Forum

Contact Info

Product

Resources

About