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.

The substantial upsurge in Web 2.0 deployment allows a large number of consumers to share their opinions with potential consumers and producers through product and service review platforms. Customer decision-making relies on an important aspect of review: spam-free opinions. Unfortunately, fraudulent activities produced spam reviews that misled potential buyers and traders. Subsequently, it does stop opinion-mining techniques from reaching accurate conclusions. This study aims to identify and review existing state-of-the-art methodologies for three groups: (1) spam reviews, (2) individual spammers, and (3) group spam. The machine learning (ML) and deep learning (DL) techniques for spam detection are categorized, and we establish an assessment that may be deemed appropriate in the field. The findings reveal a total of 10 metrics, with accuracy being the most often used (25%) concerning ML-based techniques in spam detection. Followed by recall in 24% of studies, and precision in 22% of studies. In addition, the F-measure, the area under the curve (AUC), and F1-score evaluation metrics reveal that the use of the Amazon dataset as a whole increased by 7%. This study concludes that the majority of SMS spam filtering strategies are leading solutions. In addition, the taxonomy of existing state-of-the-art methodologies is developed, and it is concluded that a substantial number of studies utilize these existing SMS anti-spam applications. This research uncovered previously unexplored areas of ML and DL’s application to spam review and provided a new paradigm for applying these technologies to the issue. The findings provide both academics and practitioners with a deeper understanding of the barriers to spam review identification, as well as potential improvement opportunities using ML techniques. These individuals will now have the chance to consider a benchmark for future research.

The automatic speech identification in Arabic tweets has generated substantial attention among academics in the fields of text mining and natural language processing (NLP). The quantity of studies done on this subject has experienced significant growth. This study aims to provide an overview of this field by conducting a systematic review of literature that focuses on automatic hate speech identification, particularly in the Arabic language. The goal is to examine the research trends in Arabic hate speech identification and offer guidance to researchers by highlighting the most significant studies published between 2018 and 2023. This systematic study addresses five specific research questions concerning the types of the Arabic language used, hate speech categories, classification techniques, feature engineering techniques, performance metrics, validation methods, existing challenges faced by researchers, and potential future research directions. Through a comprehensive search across nine academic databases, 24 studies that met the predefined inclusion criteria and quality assessment were identified. The review findings revealed the existence of many Arabic linguistic varieties used in hate speech on Twitter, with modern standard Arabic (MSA) being the most prominent. In identification techniques, machine learning categories are the most used technique for Arabic hate speech identification. The result also shows different feature engineering techniques used and indicates that N-gram and CBOW are the most used techniques. F1-score, precision, recall, and accuracy were also identified as the most used performance metric. The review also shows that the most used validation method is the train/test split method. Therefore, the findings of this study can serve as valuable guidance for researchers in enhancing the efficacy of their models in future investigations. Besides, algorithm development, policy rule regulation, community management, and legal and ethical consideration are other real-world applications that can be reaped from this research.

In the past few years, the Subjectivity and Sentiment Analysis (SSA) has garnered a lot of attention, and many unsuccessful attempts were made for developing an SSA model for the morphologically-rich languages like Arabic. In this study, we aimed to fill the gap by designing a new manual and auto-annotated corpus of the Sudanese-Dialect Arabic (SDA) along with a novel polarity lexicon. The corpus is a collection of political tweets, which were annotated by Twitter on a sentence-level. We have also described some automated, online SSA-tagging tools, which could explore all annotated data. Furthermore, we have investigated the effect of various pre-processing techniques on the Sudanese SSA process. Here, we presented a novel SSA technique for the Sudanese dialect, which displayed a good subjectivity classification result and showed an 83.5% accuracy for the Decision Tree performance.

Forecasting disease outbreaks in real-time using time-series data can help for the planning of public health interventions. We used a support vector machine (SVM) model using epidemiological data provided by Johns Hopkins University Centre for Systems Science and Engineering (JHU CCSE), World Health Organization (WHO), and the Centers for Disease Control and Prevention (CDC) to predict upcoming records before the WHO made an official declaration. Our study, conducted on the time series data available from 22 January till 10 March 2020, revealed that COVID-19 was spreading at an alarming rate and progressing towards a pandemic. The initial insight that confirmed COVID-19 cases were increasing was because these received the highest number of effects for our selected dataset from 22 January to 10 March 2020, i.e., 126,344 (64%). The recovered cases were 68289 (34%), and the death rate was around 2%. Moreover, we classified the tweets from 22 January to 15 April 2020 into positive and negative sentiments to identify the emotions (stress or relaxed) posted by Twitter users related to the COVID-19 pandemic. Our analysis identified that tweets mostly conveyed a negative sentiment with a high frequency of words for #coronavirus and #lockdown amid COVID-19. However, these anxiety tweets are an alarm for healthcare authorities to devise plans accordingly.