A Novel Stacked Ensemble for Hate Speech Recognition

Aljero, Mona Khalifa A.; Dimililer, Nazife

doi:10.3390/app112411684

Cited by 18 publications

(8 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many proposed works performed hate speech detection as a binary classification problem and considered a broad concept such as detecting bullying and derogatory language. In [23], the authors presented an original technique to detect hatred speech in English tweets. For that, they utilized three models, i.e., logistic regression (LR), XGBoost classifier (XGB), and support vector machine (SVM).…”

Section: Related Workmentioning

confidence: 99%

“…K-NN algorithm does not make any hypothesis on underlying data. Thus, it is a non-parametric algorithm [23]. KNN is a lazy learner algorithm because it holds the dataset and it achieves an activity on the dataset at classification, i.e., does not memorize from the training set immediately.…”

Section: ) K-nearest Neighbor (K-nn)mentioning

confidence: 99%

“…However, it expects if a value is True or False rather that predicting a continuous value. Linear regression fits a curve to the data while LR fit an "S" shaped logistic function [23]. Logistic regression can perform on both continuous and discrete data.…”

Section: ) K-nearest Neighbor (K-nn)mentioning

confidence: 99%

“…Then, predict the class for any given test data based on its probability to be assigned for different classes. The maximum probability determines the predicted class [23].…”

Section: ) Naive Bayes (Nb): Naive Bayes Classifier Relies Onmentioning

confidence: 99%

See 3 more Smart Citations

A Novel Machine Learning-based Framework for Detecting Religious Arabic Hatred Speech in Social Networks

Masadeh¹,

Davanager²,

Muaad³

2022

IJACSA

View full text Add to dashboard Cite

Social media platforms generate a huge amount of data every day. However, liberty of speech through these networks could easily help in spreading hatred. Hate speech is a severe concern endangering the cohesion and structure of civil societies. With the increase in hate and sarcasm among the people who contact others over the internet in this era, there is a dire need for utilizing artificial intelligence (AI) technology innovation that would face this problem. The rampant spread of hate can dangerously break society and severely damage marginalized people or groups. Thus, the identification of hate speech is essential and becoming more challenging, where the recognition of hate speech on time is crucial in stopping its dissemination. The capacity of the Arabic morphology and the scarcity of resources for the Arabic language makes the task of distinguishing hate speech even more demanding. For fast identification of Arabic hate speech in social network comments, this work presents a comprehensive framework with eight machine learning (ML) and deep learning (DL) algorithms, namely Gradient Boosting (GB), K-Nearest Neighbor (K-NN), Logistic Regression (LR), Naive Bayes (NB), Passive Aggressive Classifier (PAC), Support Vector Machine (SVM), Ara-BERT, and BERT-AJGT are implemented. Two representation techniques have been used in the proposed framework in order to extract features: a bag of words followed by BERT-based context text representations. Based on the result and discussion part, context text representation techniques with Ara-BERT and BERT-AJGT outperform all other ML models and related work with accuracy equal to 79% for both models.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: ) K-nearest Neighbor (K-nn)mentioning

confidence: 99%

Section: ) K-nearest Neighbor (K-nn)mentioning

confidence: 99%

“…Then, predict the class for any given test data based on its probability to be assigned for different classes. The maximum probability determines the predicted class [23].…”

Section: ) Naive Bayes (Nb): Naive Bayes Classifier Relies Onmentioning

confidence: 99%

See 2 more Smart Citations

A Novel Machine Learning-based Framework for Detecting Religious Arabic Hatred Speech in Social Networks

Masadeh¹,

Davanager²,

Muaad³

2022

IJACSA

View full text Add to dashboard Cite

show abstract

“…Kokatnoor et al [15] A model called Stacked Weighted Ensemble is suggested for the identification of hate speech. Along with certain separate classifiers.…”

Section: Introductionmentioning

confidence: 99%

Novel approach of Predicting Human Sentiment using Deep Learning

Shadadi

Kouser

Alamer

et al. 2022

jcse. journal. of. computer. science. and. engineering.

View full text Add to dashboard Cite

Due to its interactive and real-time character, gathering public opinion through the analysis of massive social data has garnered considerable attention. Recent research have used sentiment analysis and social media to do this in order to follow major events by monitoring people's behavior. In this article, we provide a flexible approach to sentiment analysis that instantly pulls user opinions from social media postings and evaluates them. As time passed, an increasing number of people shared their opinions on social media. More individuals can now communicate with one another as a result. Along with these advantages, it also has certain drawbacks that cause resentment in some people. Hate speech is another possibility. Hate speech impacts the community when it contains insulting or threatening language. Before it spreads, this kind of speech has to be identified and deleted from social media platforms. The process of determining whether a text's feelings reflect hatred or not involves sentiment analysis. Python language was used to analyze the Twitter dataset. There were 5000 Tweets in total in this dataset, and we used deep learning to improve the machine learning model's accuracy. The experimental outcome in both cases of the Twitter dataset uses the Random Forest approach, which has a 99 percent accuracy rate.

show abstract