Arabic sentiment analysis using dependency-based rules and deep neural networks

Diwali, Arwa; Dashtipour, Kia; Saeedi, Kawther; Gogate, Mandar; Cambria, Erik; Hussain, Amir

doi:10.1016/j.asoc.2022.109377

Cited by 12 publications

(3 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results of fine‐tuning AraBERT are shown in Table 5, the experiments include feeding AraBERT with different input formats. Since feature extraction in deep learning models is performed implicitly and can result in high accuracy without any explanation (Diwali et al, 2022). Our experiments will include training a classifier with features extracted from Information in the AraBERT decoder versus training a classifier with features extracted from Information provided by label information.…”

Section: Resultsmentioning

confidence: 99%

Enhancing Arabic‐text feature extraction utilizing label‐semantic augmentation in few/zero‐shot learning

et al. 2023

Self Cite

View full text Add to dashboard Cite

A growing amount of research use pre‐trained language models to address few/zero‐shot text classification problems. Most of these studies neglect the semantic information hidden implicitly beneath the natural language names of class labels and develop a meta learner from the input texts solely. In this work, we demonstrate how label information can be utilized to extract enhanced feature representation of the input text from a Transformer‐based pre‐trained language model such as AraBERT. In addition, how this approach can improve performance when the data resources are scarce like in the Arabic language and the input text is short with little semantic information as is the case using tweets. The work also applies zero‐shot text classification to predict new classes with no training examples across different domains including sarcasm detection and sentiment analysis using the information in the last layer of a trained classifier in a transfer learning setting. Experiments show that our approach has a better performance for the few‐shot sentiment classification compared to baseline models and models trained without augmenting label information. Moreover, the zero‐shot implementation achieved an accuracy up to 0.874 in Arabic sarcasm detection from a model trained on a sentiment analysis task.

show abstract

Section: Resultsmentioning

confidence: 99%

Enhancing Arabic‐text feature extraction utilizing label‐semantic augmentation in few/zero‐shot learning

et al. 2023

Self Cite

View full text Add to dashboard Cite

show abstract

“… Naili, Chaibi & Ben Ghezala (2017) proposed a sentiment analysis framework that incorporates Arabic dependency-based rules and deep learning models. Also, Diwali et al (2022) introduced a deep learning-based system for Arabic short answer scoring. The work aimed to provide a reliable system that can help teachers in the Arab world better use their time in other teaching activities that would increase the quality of learning in the region.…”

Section: Literature Reviewmentioning

confidence: 99%

A novel approach for Arabic business email classification based on deep learning machines

Masri

Al-Jabi

2023

PeerJ Computer Science

View full text Add to dashboard Cite

During the last decades, the reliance on email communication, especially in business, has increased significantly. Companies receive a massive amount of emails daily, that include business inquiries, customers’ feedback, and other types of emails. This inspired many researchers to propose different algorithms to classify and redistribute the numerous emails according to their content. Nowadays, emails containing Arabic text, especially in the Arab world, have raised an increasing concern since they became widely used in official correspondence. Nevertheless, just a small amount of literature focuses on Arabic text classification. Therefore, this work addresses Arabic business emails classification based on natural language processing (NLP). A dataset of 63,257 emails was used and the emails were classified as: urgency, sentiment, and topic classification. The proposed models are based on machine learning techniques and a lexicon of words on which the emails are identified. The models are composed of different settings of convolutional neural networks (CNN). A separate model was built, trained, and tested for each category. The results were promising and gave an accuracy of about 92% and a loss of less than 8%. They also proved the correctness and robustness of this work.

show abstract

“…Traditional machine learning and deep learning approaches, such as recurrent neural networks (RNN) and their variants, long short-term memory (LSTM), and the gated recurrent unit (GRU) [3], or a combination of traditional methods and deep learning models such as integrating of deep models and dependency rules [4], are frequently used to solve the sentiment analysis, ATE and APC problems. Machine learning requires handcrafted feature extraction methods.…”

Section: Introductionmentioning

confidence: 99%

Multi-Task Learning Model with Data Augmentation for Arabic Aspect-Based Sentiment Analysis

Fadel¹,

Abulnaja²,

Saleh³

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Aspect-based sentiment analysis (ABSA) is a fine-grained process. Its fundamental subtasks are aspect term extraction (ATE) and aspect polarity classification (APC), and these subtasks are dependent and closely related. However, most existing works on Arabic ABSA content separately address them, assume that aspect terms are preidentified, or use a pipeline model. Pipeline solutions design different models for each task, and the output from the ATE model is used as the input to the APC model, which may result in error propagation among different steps because APC is affected by ATE error. These methods are impractical for real-world scenarios where the ATE task is the base task for APC, and its result impacts the accuracy of APC. Thus, in this study, we focused on a multi-task learning model for Arabic ATE and APC in which the model is jointly trained on two subtasks simultaneously in a single model. This paper integrates the multi-task model, namely Local Cotext Foucse-Aspect Term Extraction and Polarity classification (LCF-ATEPC) and Arabic Bidirectional Encoder Representation from Transformers (AraBERT) as a shred layer for Arabic contextual text representation. The LCF-ATEPC model is based on a multi-head selfattention and local context focus mechanism (LCF) to capture the interactive information between an aspect and its context. Moreover, data augmentation techniques are proposed based on state-of-the-art augmentation techniques (word embedding substitution with constraints and contextual embedding (AraBERT)) to increase the diversity of the training dataset. This paper examined the effect of data augmentation on the multi-task model for Arabic ABSA. Extensive experiments were conducted on the original and combined datasets (merging the original and augmented datasets). Experimental results demonstrate that the proposed Multi-task model outperformed existing APC techniques. Superior results were obtained by AraBERT and LCF-ATEPC with fusion layer (AR-LCF-ATEPC-Fusion) and the proposed data augmentation word embedding-based method (FastText) on the combined dataset.

show abstract

Arabic sentiment analysis using dependency-based rules and deep neural networks

Cited by 12 publications

References 54 publications

Enhancing Arabic‐text feature extraction utilizing label‐semantic augmentation in few/zero‐shot learning

Enhancing Arabic‐text feature extraction utilizing label‐semantic augmentation in few/zero‐shot learning

A novel approach for Arabic business email classification based on deep learning machines

Multi-Task Learning Model with Data Augmentation for Arabic Aspect-Based Sentiment Analysis

Contact Info

Product

Resources

About