A Multi-Criteria Approach for Arabic Dialect Sentiment Analysis for Online Reviews: Exploiting Optimal Machine Learning Algorithm Selection

Abstract

A sentiment analysis of Arabic texts is an important task in many commercial applications such as Twitter. This study introduces a multi-criteria method to empirically assess and rank classifiers for Arabic sentiment analysis. Prominent machine learning algorithms were deployed to build classification models for Arabic sentiment analysis classifiers. Moreover, an assessment of the top five machine learning classifiers’ performances measures was discussed to rank the performance of the classifier. We integrated the top five ranking methods with evaluation metrics of machine learning classifiers such as accuracy, recall, precision, F-measure, CPU Time, classification error, and area under the curve (AUC). The method was tested using Saudi Arabic product reviews to compare five popular classifiers. Our results suggest that deep learning and support vector machine (SVM) classifiers perform best with accuracy 85.25%, 82.30%; precision 85.30, 83.87%; recall 88.41%, 83.89; F-measure 86.81, 83.87%; classification error 14.75, 17.70; and AUC 0.93, 0.90, respectively. They outperform decision trees, K-nearest neighbours (K-NN), and Naïve Bayes classifiers.

Heterogeneous Ensemble Deep Learning Model for Enhanced Arabic Sentiment Analysis

“…The result showed that Bidirectional Multi-Layer LSTM has high accuracy. In [ 12 ], author applied SVM, LR, DT, NB, and DL models on the Saudi dialect sentiment Arabic tweets dataset. The result shows that deep learning and SVM classifiers perform best with accuracy.…”

“…Recently, the research used ML algorithms to predict sentiment analysis and to enhance the sentiment analysis method’s accuracy [ 9 , 10 , 11 ], transit from primarily linear models and evolved until it reached what we see today of the more complex deep neural network. In addition, deep learning (DL) algorithms have been used to extract features with great effectiveness over other ML algorithms such as [ 12 , 13 ].…”

An Improved Sentiment Classification Approach for Measuring User Satisfaction toward Governmental Services’ Mobile Apps Using Machine Learning Methods with Feature Engineering and SMOTE Technique

“…The performance of each classifier was measured by computing the following performance measures: classification accuracy, precision, recall, and F1-score. These measures are commonly used to evaluate the performance of ML models in many research areas, such as rumor detection systems [43,44], clickbait detection [33], as well as in SA [45][46][47].…”

A Novel Hybrid Sentiment Analysis Classification Approach for Mobile Applications Arabic Slang Reviews

“…a) Fifth phase: Hybrid supervised classification approach phase: This phase performs two subsections: the first issue is applying ML approach which performs five selected ML classifiers which utilized extensively for ASA: Logistic Regression (LR) [25] [26] [27] [28], Naïve Bayes (NB) [29] [30] [31] [32], K-Nearest Neighbors (KNN) [33] [34] [31] [35], Random Forest (RF) [36] [37] [38] and SVM [33] [39] [40] in addition applying DL approach which performs DL classifier Multi-Layer Perceptron Neural Network (MLP-NN) which applied in [36] [37] for ASA.…”