Mazajak: An Online Arabic Sentiment Analyser

Farha, Ibrahim Abu; Magdy, Walid

doi:10.18653/v1/w19-4621

Cited by 87 publications

(94 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [38], the authors used an ensemble approach to combine the performance of a CNN with an LSTM model using the ASTD dataset with three sentiment classes: positive, negative and neutral. The accuracy produced by the CNN model was 64.3%, the accuracy produced by LSTM was 64.75%, and the combined ensemble accuracy is 65.05%, which was much lower than the results produced by other papers using the same data (for example, the accuracies produced in [36]). The authors used pretrained word embeddings to train the models.…”

Section: Literature Review and Related Work: Models And Datasetscontrasting

confidence: 56%

See 1 more Smart Citation

A Novel Deep Learning-Based Multilevel Parallel Attention Neural (MPAN) Model for Multidomain Arabic Sentiment Analysis

2021

View full text Add to dashboard Cite

Section: Literature Review and Related Work: Models And Datasetscontrasting

confidence: 56%

“…In [36], the authors applied a CNN combined with long short-term memory (LSTM) to analyze sentiments in three public datasets: SemEval 2017, ASTD, and ARSAS. They used the word2vec model to generate the required word embeddings.…”

Section: Literature Review and Related Work: Models And Datasetsmentioning

confidence: 99%

A Novel Deep Learning-Based Multilevel Parallel Attention Neural (MPAN) Model for Multidomain Arabic Sentiment Analysis

2021

View full text Add to dashboard Cite

“…The reason for the difference in accuracy should be attributed to the insufficient number of samples in the training set, and the quality is also a factor limiting the accuracy. In terms of recall, the Bi-LSTM model is superior to the CNN-LSTM model [58] on an average level. In terms of F1-Score, the Bi-LSTM model far exceeds the two types of models proposed by the papers [59] [60].In summary, compared with the SOTA, the Bi-LSTM model proposed in this paper has a certain gap in some indicators, such as accuracy and recall.…”

Section: Comparative Experiments With Sotamentioning

confidence: 92%

Improved Danmaku Emotion Analysis and Its Application Based on Bi-LSTM Model

et al. 2020

View full text Add to dashboard Cite

With the rapid development of social media, danmaku video provides a platform for users to communicate online. To some extent, danmaku video provides emotional timing information and an innovative method to analyze video data. In the age of big data, studying the characteristics of danmaku and its emotional tendencies can not only help us understand the psychological characteristics of users but also feedback the effective information of users to video platforms, which can help the platforms optimize related short video recommendations so that it can provide a more accurate solution for the selection of audiences during video production. However, danmaku is different from traditional comments. Current emotion classification methods are only suitable for two-dimensional classification which are not suitable for danmaku emotion analysis. Aiming at the problems such as the colloquialism, diversity, spelling errors, structural non-linearity informal language on the Internet, diversity of social topics, and context dependency of emotion analysis of the danmaku data, this paper proposes an improved emotion analysis model based on Bi-LSTM model to classify the further four-dimensional emotions of Pleasure, Anger, Sorrow and Joy. Furthermore, we add tags such as comment time and user name to the danmaku information. Experimental results show that the improved model has higher Accuracy, Recall, Precision, and F1-Score under the same conditions compared with the CNN and SVM. The classification effect of improved model is close to the SOTA. Experimental results also show that the improved model can be effectively applied to the analysis of irregular danmaku emotion.

show abstract

“…It is worth to mention that for PBLM and HATN, we have used an extra 4000 unlabeled sentences from each domain/dialect. For HTAN, we have used Mazjak word embedding model (Abu Farha and Magdy, 2019)…”

Section: Compared Methodsmentioning

confidence: 99%

Domain Adaptation for Arabic Cross-Domain and Cross-Dialect Sentiment Analysis from Contextualized Word Embedding

Mekki¹,

Mahdaouy²,

Berrada³

et al. 2021

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

Finetuning deep pre-trained language models has shown state-of-the-art performances on a wide range of Natural Language Processing (NLP) applications. Nevertheless, their generalization performance drops under domain shift. In the case of Arabic language, diglossia makes building and annotating corpora for each dialect and/or domain a more challenging task. Unsupervised Domain Adaptation tackles this issue by transferring the learned knowledge from labeled source domain data to unlabeled target domain data. In this paper, we propose a new unsupervised domain adaptation method for Arabic cross-domain and crossdialect sentiment analysis from Contextualized Word Embedding. Several experiments are performed adopting the coarse-grained and the fine-grained taxonomies of Arabic dialects. The obtained results show that our method yields very promising results and outperforms several domain adaptation methods for most of the evaluated datasets. On average, our method increases the performance by an improvement rate of 20.8% over the zero-shot transfer learning from BERT.

show abstract

Mazajak: An Online Arabic Sentiment Analyser

Cited by 87 publications

References 30 publications

A Novel Deep Learning-Based Multilevel Parallel Attention Neural (MPAN) Model for Multidomain Arabic Sentiment Analysis

A Novel Deep Learning-Based Multilevel Parallel Attention Neural (MPAN) Model for Multidomain Arabic Sentiment Analysis

Improved Danmaku Emotion Analysis and Its Application Based on Bi-LSTM Model

Domain Adaptation for Arabic Cross-Domain and Cross-Dialect Sentiment Analysis from Contextualized Word Embedding

Contact Info

Product

Resources

About