Search

In recent times, there has been a notable surge in the interest of researchers in the realm of document analysis and optical character recognition (OCR). Significant advancements have been made in OCR engines across various languages, encompassing both printed and handwritten documents. However, there has been a comparatively lower focus on processing documents written in Arabic when juxtaposed with languages like English. This discrepancy arises from several factors, including the inherent challenges posed by the Arabic language and the limited availability of Arabic document datasets. To implement any OCR engine, the initial step involves analyzing the layout of images before subjecting them to the OCR process. This thesis specifically delves into the realm of layout analysis for historical Arabic documents, employing a deep learning (DL) approach. The chosen methodology utilizes the Mask Region-based Convolutional Neural Network (RCNN). The dataset employed consists of historical Arabic documents, particularly early printed ones, each characterized by unique sizes, structures, and processing prerequisites. Processing historical documents is inherently more challenging due to factors such as the document's layout structure, distinctive handwriting styles of the authors, paper aging, historical timeframe, ink properties, and more. The achieved accuracy result is 51.14%. When juxtaposed with other existing models, it becomes evident that this work attains a state-of-the-art status, showcasing an impressive outcome. [ABSTRACT FROM AUTHOR]

Topic Modeling approaches face difficulties in processing legal texts because of their unique characteristics, such as the length of the texts and the specialized terminology used within them. The process of topic modeling involves finding a text's semantic structure. This way, specific approaches are needed. When the legal documents are presented has a lot to do with what topics are important. This paper aims to explain and evaluate BERTopic's application to topic modeling in legal documents. In this research, we experiment with BERTopic by utilizing its several pre-trained Arabic language models as embeddings. Performance evaluation employs the Normalized Pointwise Mutual Information (NPMI) measure. Notably, in comparison to multilingual pre-trained models, our findings reveal that BERTopic using Arabic monolingual pre-trained models exhibits superior performance, offering insights into sustainable and efficient topic modeling for legal documents. [ABSTRACT FROM AUTHOR]

The recognition of sign language has significant importance as an application that addresses the communication challenges faced by those within the Deaf and hard-of-hearing populations. This study introduces a methodology for the identification of sign language using spatio-temporal attributes, specifically concentrating on two distinct sign language datasets: the CSL (Chinese Sign Language) alphabet and the ArSL (Arabic Sign Language) alphabet. The study's objective is to create a powerful recognition system capable of reliably identifying sign language motions. A combination of spatio-temporal characteristics collected from video sequences of sign language motions is used in the suggested technique. These characteristics capture the spatial layout as well as the temporal dynamics of the motions, allowing the model to recognise signals with more precision. Experiments were carried out on the CSL and ArSL alphabet datasets to validate the method's efficacy. The findings of the trials are encouraging. The suggested system recognized CSL alphabet signs with an accuracy of 90.87% and ArSL alphabet signs with an accuracy of 89.46%. These high accuracy rates show the power of the spatiotemporal feature-based method for sign language identification. The system's success on two different sign languages implies that it is adaptable and useful in a wide variety of sign language applications. This study contributes to the evolution of assistive technology, making sign language recognition more accessible and efficient for those who communicate using sign language. Furthermore, the findings pave the way for additional research into spatiotemporal feature-based approaches in sign language recognition, with potential applications in real-world settings such as sign language interpretation and communication support. [ABSTRACT FROM AUTHOR]

This paper examines the utilization of Arabic tautological expressions, provided by customers as a reiteration of an element of an English compound such as 'apple pie' [ fatˁi:rit ʔabil bai 'an apple pie pie'], from the perspective of Relevance Theory (RT). Data was collected from 90 participants, divided into two groups: 40 native speaker informants recruited from various shopping malls, and 50 shop assistants in Amman, Jordan. The first group was interviewed, whereas the second group filled out a multiple-choice task aimed at identifying the type of expression they typically receive when taking orders from customers. The data analysis revealed that Arabic tautological expressions are more frequently used compared to other alternatives. Drawing on RT, we argue that Arabic tautological expressions function as an ostensive stimulus. We propose that tautological expressions, utilized within a code-switching context, function as a means of pragmatic enrichment. This occurs through the process of ad hoc construction, wherein speakers refine the originally encoded concept to convey corrective or contrasting meanings. Instead of lacking information, the use of tautological phrases can be seen as a pragmatic tactic to maximize relevance by minimizing cognitive effort, optimizing successful communication of the intended meaning, and reinforcing contextual effects. • Examining the use of Arabic tautology in English compounds provided by customers. • Arabic tautologies are preferred and more frequently used than alternatives. • Drawing on RT, Arabic tautological expressions function as an ostensive stimulus. • The use of tautology conveys corrective or contrastive meanings via ad hoc concept construction. • The use of tautological phrases can be seen as a pragmatic tactic. [ABSTRACT FROM AUTHOR]

Moving beyond classical single-label classification tasks proves to be crucial nowadays since the majority of real-life problems are inherently multi-label. Various tasks require involving, for one single instance as an input, several classes simultaneously or none of them. Particularly, detecting abusive texts in Arabic social networks intuitively carries several manifestations with regard to the targets' referred characteristics like gender, race, sexual orientation, etc. Existing multi-label algorithms either invest considerable complexity to handle labels' correlation or explicitly ignore this information to deal with each label as an independent binary problem. To remedy the abovementioned problems, this paper proposes a new multi-label classification solution based on deep neural networks. The main idea is to exploit the correlation between all the dataset's labels to break it into ordered subsets. Then, follow that order to implement a chain of BERT classifiers. Each classifier is therefore trained on a corresponding subset and passes its output via feature space to the following one. For a new instance, we gather all the classifiers' results and pick the best one for each class. Experiments showed that our approach significantly outperformed the single classifier's method using BERT as well as other neural network architectures on our Arabic abusive language dataset. 91%,76%, and 82% were achieved respectively for precision, recall, and F1-score. Moreover, we further investigated the performance of our approach on Kaggle multi-label research paper dataset and achieved very promising results. [ABSTRACT FROM AUTHOR]

Large Language Models (LLMs) have ushered in a major paradigm shift in Natural Language Processing (NLP), where large pre-trained Language models (LMs) have become a fundamental component of most NLP tasks. These models are intelligent enough to find relevant and meaningful representations of a language without any supervision. They are used to fine-tune typical NLP tasks with substantially higher precision than conventional shallow learning techniques. However, training these models requires a massively large corpus that adequately represents a language. Due to the availability of enormous corpora, English LLMs typically perform better than their counterparts. This effort focuses on the design and development of a large Arabic corpus. The corpus comprises over 500 GB of Arabic cleaned text, intended to improve cross-domain knowledge and downstream generalization capability of LLMs. The corpus was employed in the training of a large Arabic LLM. In order to assess the efficacy of the LLM, a variety of typical NLP tasks were fine-tuned. The fine-tuned tasks exhibited a significant boost in accuracy ranging between 4.5 and 8.5%, when compared to those downstreamed from multi-lingual BERT (mBERT). To the best of our knowledge, this is currently the largest clean and diverse Arabic corpus ever assembled. [ABSTRACT FROM AUTHOR]

Arabic affect analysis on Twitter avidly helps to capture the emotional states of individuals being expressed regarding many targets, such as world-level events, products, and services. It is the key to monitoring and advancing human intelligence, which impacts human decision-making processes efficiently. However, state-of-the-art models have not witnessed serious developments yet since they have just achieved an accuracy of around 54%. This inaccuracy is mainly due to the agglutination, dialectal variation, and morphological richness of the Arabic language, as well as the unique features of tweets, such as shortness, noisiness, and informal language. This paper presents an approach that tackles these challenges and then improves the performance of Arabic affect analysis on Twitter. First, we propose a novel feature-based fusion representation for Arabic tweets to capture the polysemy, semantic/syntactic information, and conveyed emotional knowledge; and to deal with Out-Of-Vocabulary (OOV) words. Second, we propose an attentional deep learning model based on Bidirectional Gated Recurrent Unit (BiGRU), Bidirectional Long Short-Term Memory (BiLSTM), and Convolution Neural Network (CNN) to effectively learn local and global features and provide a multilabel emotional classification. The experimental results indicate that our proposal outperforms twelve state-of-the-art and baseline methods with a significant improvement of 6% in accuracy. [ABSTRACT FROM AUTHOR]

Nowadays, sentiment analysis has become a vital research field involving various research fields, including data mining, computational linguistics, and natural language processing. It seeks to identify and extract the sentiment polarity of written textual data (post, tweet, review, etc.). Context-sensitive representations produced by the language model ELMo are generally aided with the polysemy and capturing words with diverse sentiment valence, allowing for more effective handling of the Arabic language's ambiguity and complexity. Recently, deep learning has shown robust data modelling capabilities for coping with complex and massive datasets. In various language models, it is considered the most advanced-based model. In this work, firstly, we investigate the effectiveness of deep learning models with context-sensitive representations. Finally, we develop a bidirectional LSTM deep model for sentiment analysis on Arabic book reviews with the ability to capture the polysemy in the context. This study aims to determine whether book reviews can be categorised as displaying positive or negative sentiments. The experiments prove that our proposed method considerably outperforms the state-of-the-art sentiment analysis tailored to Arabic book reviews. [ABSTRACT FROM AUTHOR]

Pronunciation feedback is essential for teaching languages to children, urging the need to create computer-assisted pronunciation training (CAPT) systems to automate this process. Most of the current CAPT systems for Arabic Language examined a wide age range, including children, with a dominant adults' portion. However, these systems did not consider many challenges facing children's voice, in terms of the acoustic difference between children and adults. Moreover, due to the lack of publicly Arabic datasets, Arabic-related systems were evaluated on datasets consisting of the sound of Arabic letters, with minor efforts to contain words or sentences. In this paper, we propose the Arabic Utterance Mispronunciation Detector for Children (AUMD-Child) system that detects the mispronunciation of children for Arabic words at real-time. The proposed system is based on the fusion of Vision Transformer (ViT) and transfer learning-based model (AlexNet). A novel public dataset is constructed, in which 16 words have been collected from children aged between 7 and 12 years old to evaluate our proposed system. The experimental results, compared to the state-of-the-art models in literature, showed that the accuracy of our proposed AUMD-Child system exceeded the handcrafted features-based models by an average of 7 %, the CNN features-based models using AlexNet and SVM by 5 %, transfer learning-based model using AlexNet by 4 %, and transformers-based model using ViT by 2 %, achieving an average accuracy of 91.81 % for detecting the mispronunciation patterns in Arabic words but in an average processing time of 33 ms, exceeding the other state-of-the-art models with 13 %, 0.03 %, 0.06 %, and 23 % respectively. • Proposed AUMD-Child system as first Arabic CAPT to detect children mispronunciation. • Exclusively utilize Vision Transformer in Arabic CAPT, fused with transfer learning. • Conducted comparative experimental evaluation with state-of-the-art for children. • Introducing a new Arabic words dataset for children aged between 7 and 12 years old. [ABSTRACT FROM AUTHOR]

Deaf-mute individuals encounter substantial difficulties in their daily lives due to communication impediments. These individuals may encounter difficulties in social contact, communication, and capacity to acquire knowledge and engage in employment. Recent studies have contributed to decreasing the gap in communication between deaf-mute people and normal people by studying sign language interpretation. In this paper, a real-time Arabic avatar system is created to help deaf-mute people communicate with other people. The system translates text or spoken input into Arabic Sign Language (ArSL) movements that the avatar makes using deep-learning-based translation. The dynamic generation of the avatar movements allows smooth and organic real-time communication. In order to improve the precision and effectiveness of ArSL translation, this study depends on a state-of-the-art deep learning model, which makes use of YOLOv8, to recognize and interpret sign language gestures in realtime. The avatar is trained on three diverse datasets of Arabic sign language images, namely Sign-language-detection Image (SLDI), Arabic Sign Language (ArSL), and RGB Arabic Alphabet Sign Language (AASL), enabling it to accurately capture the nuances and variations of hand movements. The best recognition accuracy of the suggested approach was 99.4% on the AASL dataset. The experimental results of the suggested approach demonstrate that deaf-mute people will be able to communicate with others in Arabic-speaking communities more effectively and easily. [ABSTRACT FROM AUTHOR]

• Detecting Arabic fake news is a difficult task due to the lack of datasets. • "ArabicFakeNews" is an Arabic fake dataset that we compiled manually from different Sources. • Proposed models are based on deep learning and contextual pre-trained Models. • AraBERT-v02 model shows best result over other pretrain models and several baselines. Fake news has become a serious problem due to many reasons; its rapid spread across the Internet, the difficulty of discovering and distinguishing it from the real news, and the increasing dependence of individuals on social media platforms as the main source. In addition, it has harmful consequences at different levels; individual, community, political, and financial levels. Arabic fake news detection task still needs a considerable amount of effort, due to the lack of datasets and limited research in this field. In this thesis, we investigate detecting the credibility of news in Arabic, based on deep learning methods and transformers models for capturing the hidden information and pattern rather than depending on a set of handcrafted features, that have been addressed in most of the related work in the literature. Also, we release "ArabicFakeNews" which is a manual fake news dataset that contains 2 k Arabic fake news collected from different sources. We also study how the length of news can affect the model performance by applying the deep learning model and transformer-based model on three datasets with different average lengths (title, description, and text dataset). The description dataset performs better in classifying the news, this is due to the fact that the average length of fake news is close to the average length of real news in this dataset. The evaluation result shows that transformers models outperform the deep learning models on the three datasets. The transformer results show that AraBERTv2 on the "description" dataset gets the highest result among all other transformers with an accuracy of 0.97, 0.97 f1-sore, 0.97 precision, and 0.9658 recall. [ABSTRACT FROM AUTHOR]

This paper addresses the potential of Arabic Sign Language (ArSL) recognition systems to facilitate direct communication and enhance social engagement between deaf and non-deaf. Specifically, we focus on the domain of religion to address the lack of accessible religious content for the deaf community. We propose a multimodal architecture framework and develop a novel dataset for ArSL production. The dataset comprises 1950 audio signals with corresponding 131 texts, including words and phrases, and 262 ArSL videos. These videos were recorded by two expert signers and annotated using ELAN based on gloss representation. To evaluate ArSL videos, we employ Cosine similarities and mode distances based on MobileNetV2 and Euclidean distance based on MediaPipe. Additionally, we implement Jac card Similarity to evaluate the gloss representation, resulting in an overall similarity score of 85% between the glosses of the two ArSL videos. The evaluation highlights the complexity of creating an ArSL video corpus and reveals slight differences between the two videos. The findings emphasize the need for careful annotation and representation of ArSL videos to ensure accurate recognition and understanding. Overall, it contributes to bridging the gap in accessible religious content for deaf community by developing a multimodal framework and a comprehensive ArSL dataset. [ABSTRACT FROM AUTHOR]

Speech is one of the attributes that humans enjoy. Humans possess several distinguishing characteristics, including fingerprints, hands, fingers, eyes and DNA, which set each individual apart from one another. Numerous datasets specialise in speech and include many languages from around the world, such as JNAS DATASET, TCDSA DATASET, UF-VAD DATASET, NIST SRE DATASET, a Gender Dataset, TIMIT dataset, common-voice dataset, SITW, VoxCeleb1, VoxCeleb2, Arabic-Saudi Arabic Speech Dataset-2 and MGB Challenge Dataset. This research contributes to the establishment of requirements for creating a new dataset featuring Arab celebrities in the Arabic language. The Hu-Int dataset is designed with the goal of profiling new notable Arab individuals. It includes 1017 speakers after the filtering process and 93.814 videos with various formulas (e.g. mp4, wav and CSV). The algorithms Random Forest, Logistic Regression, SVM (RBF), SVM (Linear), Decision Tree and CNN-LSTM are then used. The initial experimental results for algorithmic precision yielded the following accuracy rates: 0.88%, 0.87%, 0.96%, 0.77%, 0.85% and 0.88%, respectively, for males and 0.86%, 0.84%, 0.92%, 0.79%, 0.87% and 0.89%, respectively, for females in gender classification. The features of age were divided into six classes for gender classification. Dimensionality reduction was performed using principal component analysis and linear discriminant analysis. The results showed the opposite effect, with the SVM (RBF) and hybrid CNN-LSTM algorithms outperforming many other algorithms in both gender detection and age estimation. [ABSTRACT FROM AUTHOR]

• This study explores diglossic/multiglossic settings, encompassing oral and sign languages. • Significant language disparities are observed between deaf and hearing Arab children. • Hearing children exhibit longer and more complex narratives in the study. • Hearing children prefer Standard Arabic, while deaf children show diverse language forms. • Deaf children's narratives reflect varied language exposures and environments. This study explores the narrative skills of deaf and hearing children within the context of Arabic diglossia, a linguistic environment characterised by significant differences between spoken dialects and formal written language. Using Stein and Glenn's (1979) and Bruner's (1991) frameworks, the research analyses the narrative constructions of 13 hearing and 13 deaf children in Kuwait. The findings reveal that hearing children, benefiting from consistent exposure to spoken and formal Arabic, produced more coherent and detailed narratives compared to deaf children. Hearing participants also demonstrated greater vocabulary diversity. Age-related improvements in narrative skills were more pronounced among hearing children, while the impact of sign language exposure on narrative abilities was significant among deaf children. The study underscores the critical role of early language exposure and educational support in fostering narrative development, particularly in a diglossic context. These findings highlight the need for specialised educational strategies to support the unique narrative development needs of deaf children. [ABSTRACT FROM AUTHOR]

Dyslexia is one of the most common learning disabilities. It poses challenges in reading skills development and can have a lasting impact in other areas of learning. To investigate the prevalence of developmental dyslexia among primary school children aged 6–13 in Arab countries, and to explore associations with gender and residency. Following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines, we included studies up to December 2023 that were found in English and Arabic research database searches. Random-effects proportional meta-analyses determined the prevalence of dyslexia in Arab countries, with a direct comparison between girls and boys using odds ratios. The meta-analysis included 18 studies with 30,243 participants: eight studies (44 %) from Egypt, four (22 %) from Saudi Arabia, and six (34 %) from other Arab countries. The pooled prevalence of dyslexia for children aged 6–13 in these studies was 11 %. The odds ratio for girls vs. boys was 0.85 (95 %). Geographical variations were notable, with eight studies (44 %) from Asia and ten (56 %) from Africa. Eight studies (44 %) were from the Gulf region, while ten (56 %) were from non-Gulf region. Prevalence of dyslexia was significantly higher in Arab countries in Asia (24 %) than in Africa (12 %), and the prevalence in non-Gulf countries was 13 % which was nearly half in comparison to Gulf countries (24 %). The prevalence rates of dyslexia in primary school children varies across the Arab world, with higher rates among boys, Arab countries in Asia, and Gulf countries. Further research is needed to include both genders and standardized diagnostics across more Arab countries to understand and address dyslexia effectively. • Dyslexia affects 11 % of Arab primary students, with higher rates among boys. Gulf region shows higher prevalence (24 %) than non-Gulf (13 %). • Meta-analysis (18 studies, 30,243 participants) shows dyslexia prevalence higher in Arab countries in Asia (24 %) than in Africa (12 %). • Research needs standardized diagnostics in Arab countries to address dyslexia, considering gender differences and Arabic language challenges. [ABSTRACT FROM AUTHOR]

NLP resources play a crucial role in the building of many NLP applications. The importance of these resources depends not only on their size and coverage but also on the richness and the precision of the annotated information they provide. In the case of resource-scarce languages such as Moroccan Arabic, the building of NLP applications is limited due to the lack of these resources. To overcome this problem, we follow a rule-based approach to generate a Moroccan morphological vocabulary (MORV) which constitutes the first step addressing the problem of Moroccan morphological generation. MORV is designed and implemented based on two main components: On one hand, an MA lexicon and a list of fully annotated affixes and clitics that we have created specifically to ensure the generation process. On the other hand, a set of rules covering the concatenation and the orthographic adjustments of the generated words. Moreover, given a base form, MORV outputs more than 4.5 M Moroccan words with rich morphological features such as tense, gender, number, state, etc. We tested the coverage of MORV on texts collected from Moroccan social media and realized that it reaches a vocabulary coverage of 84% and a precision of 94%. This system is a benefit for building other NLP applications such as spell checking, morphological analysis, and machine translation. [ABSTRACT FROM AUTHOR]

With the widespread growth of data on the Web, the need for efficient methods to get and arrange valuable information from these big noisy data is increased. The knowledge graph (KG) is a way to represent and organize the data in a more efficient and easy way to modify, use, and understand. Recently, KG has become a new hotspot topic in academic and business research; it is used in many applications such as intelligent question-answering (QA), recommender systems, map navigation, etc. There has been a tendency to construct KG in different languages such as English, Chinese, Persian, or Arabic. Constructing KG faces many challenges and obstacles, especially constructing Arabic Knowledge Graph (AKG) due to the sparse Arabic data in online encyclopedias and academic research, as well as the lack of tools that can process the proprietary nature of the Arabic language effectively, besides other challenges. This research aims to review and discuss KG construction best practices (systems, phases, problems, and challenges) with highlighting the Arabic perspective. Besides, it elaborates a classification of the AKG challenges and investigates the potential solutions and opportunities that might define the key future research directions of constructing AKG. [ABSTRACT FROM AUTHOR]

Superior stemming algorithms aid significantly in many natural language processing (NLP) applications such as information retrieval. Arabic light-based stemmer is one of the most important stemming algorithms. However, partially due to the highly inflected and complexity of Arabic language morphological structure, most of the existing Arabic light-based stemmer algorithms eliminate a few numbers of suffixes and prefixes or both in the process of recognising the infix patterns to determine roots. The elimination of suffixes and prefixes leads to many inefficient results. Hence, this study aims to develop an improved light-based algorithm of the Arabic stemmer by proposing an appropriate suffixes and prefixes list, which is supported by rules according to word length (without using a morpheme or patterns on a stem). Our improved Dlight Arabic stemmer focuses on determining and removing the infix patterns under many rules on length-words and according to a specific order of the stages of the stemming to extract the double, triple and quadruple roots from long and short Arabic words. To evaluate our proposed light-based Arabic stemmer, we compared our stemmer against existing Arabic stemmers, namely Light10, Condlight and ARLST. The experimental results showed the proposed Develop Arabic Light-Based Stemmer (Dlight) obtained the best performance with 68% of F-measure, while the other three Arabic stemmers yield slightly lower F-measure. Finally, establishing an appropriate list of suffixes and prefixes with word length rules to stem Arabic words can improve the performance of a light-based Arabic stemmer. [ABSTRACT FROM AUTHOR]

One of the most important phases in text processing is stemming, whose aim is to aggregate all variations in a word into one group to aid natural language processing. The morphological structure of the Arabic language is more challenging than that of the English language; thus, it requires superior stemming algorithms for Arabic stemmers to be effective. One of the challenges is the irregular broken plural, which has been a problematic issue in Arabic natural language processing that affects the performance of Arabic information retrieval and other Arabic language engineering applications. Several studies have attempted to develop solutions to irregular plural problems, but the challenge remains, especially in extracting correct Arabic root words. In this paper, the broken plural rule (BPR) algorithm introduces new solutions to solve the problem in which an existing root-based method cannot extract correct roots by using their proposed rules. The BPR algorithm introduces several rules (main rules and subrules) to extract the correct roots of the Arabic irregular broken plural words. To evaluate the effectiveness of the BPR algorithm, we extracted roots from an Arabic standard dataset and applied the BPR algorithm as an enhancement to a root-based Arabic stemmer, ISRI. The obtained results from both evaluations showed encouraging results: (i) Only a few numbers of incorrect roots were stemmed on the large-sized Arabic word dataset. (ii) The enhanced root-based Arabic stemmer, ISRI + BPR, exhibited the best performance compared with the original ISRI stemmer and a well-known Arabic stemmer, ARLS 2. Thus, the proposed BPR algorithm has solved some of the irregular broken plural problems that eventually increase the performance of a root-based Arabic stemmer. [ABSTRACT FROM AUTHOR]

Reading traffic signs while driving a car for visually impaired people and people with visual problems is a very difficult task for them. This task is encountered every day, sometimes incorrect reading of traffic signs can lead to very serious results. In particular, the Arabic language is very difficult, making recognizing and viewing Arabic text a difficult task. In this context, we are looking for an effective solution to remove errors and results that can sometimes end someone's life. This article aims to correctly read traffic signs with Arabic text using augmented reality technology. Our system is composed of three modules. The first is text detection and recognition. The second is Text visualization. The third is Text to speech methods conversion. With this system, the user can have two different results. The first result is visual with much-improved text and enhancement. The second result is sound, he can hear the text aloud. This system is very applicable and effective for daily life. To assess the effectiveness of our work, we offer a survey to a group of visually impaired people to give their opinion on the use of our application. The results have been good for most people. [ABSTRACT FROM AUTHOR]

Readability is a measure that associates a written text to a reader's skill or grade level. Readability assessment is very important in the field of second or foreign language (L2) learning. The complexity of reading in some languages, such as Arabic, presents a great challenge to overcome and increases the need for such a measure before giving texts to a learner. In order to develop powerful tools that automate this choice, we must first choose the optimal feature set to be used in the development of such a tool. In this paper, we present an approach to automatically measure the readability of Arabic as a foreign language through a series of experiments. We begin by using a wide range of features plausibly relevant to readability, as found in the literature, and reduce them in subsequent experiments by eliminating features that appear to have little significance in readability prediction. Our objective is to keep the smallest set of features that gives good readability prediction accuracy based on three different corpora for learners of Arabic as a foreign language, annotated with five difficulty levels. Our best L2 readability accuracy result is 86.15%. [ABSTRACT FROM AUTHOR]

Update summarization is a relatively recent summarization task concerned with creating a short summary from news articles, assuming the user has already read a number of previous articles. It is useful for users who want to know the latest information about a specific topic. The availability of systems that provide update summaries saves user time and effort. Unfortunately, such resources are lacking for the Arabic language. This paper aims to provide an update summarization system that generates an update summary containing the latest information requested by a user from multiple documents. This paper provides an Arabic query-based Update-Summarization System (AUSS). We use a graph-based ranking model to represent the similarity, through a combination of lexical and semantic relations between words. Our experiments show that AUSS achieves promising results, achieving a best F-Measure of 0.5405. To test AUSS, we created a new corpus especially for Arabic Update Summaries, which we call (AUS-DB). AUS-DB contains 183 articles and their corresponding reference summaries. [ABSTRACT FROM AUTHOR]

Abstractive text summarisation is essential to producing natural language summaries with main ideas from large text documents. Despite the success of English language-based abstractive text summarisation models in the literature, they are limitedly supporting the Arabic language. Current abstractive Arabic summarisation models have several unresolved issues, a critical one of which is syntax inconsistency, which leads to low-accuracy summaries. A new approach that has shown promising results involves adding topic awareness to a summariser to guide the model by mimicking human awareness. Therefore, this paper aims to enhance the accuracy of abstractive Arabic summarisation by introducing a novel topic-aware abstractive Arabic summarisation model (TAAM) that employs a recurrent neural network. Two experiments were conducted on TAAM: quantitative and qualitative. Based on a quantitative approach using ROUGE matrices, the TAAM model achieves 10.8% higher accuracy than other existing baseline models. Additionally, based on a qualitative approach that captures users' perspectives, the TAAM model is capable of producing a coherent Arabic summary that is easy to read and captures the main idea of the input text. [ABSTRACT FROM AUTHOR]

Research work in the field of offline Arabic handwriting recognition has seen an exponential growth during past decades. Despite much work dedicated to the recognition of handwritten Arabic text, there remains a lot to achieve in this regard. Though there exists a lot of work that is confined to the Arabic text, this paper presents an approach towards classifying and recognizing text written in one of the famous scripts of Arabic language i.e. Kufic script. The approach based on character segmentation does not perform well in recognizing Kufic text due to various complexities. The proposed system is based on word segmentation and employs a Histogram of Oriented Gradient (HOG) and Local Binary Pattern (LBP) feature extraction techniques. Later in the paper a comparison is given between the numerical results of proposed technique with previous Arabic text recognition techniques to show the effectiveness of present work. This approach yields effective results with 97.05% accuracy in recognizing the Arabic text written in Kufic script using Polynomial kernel of SVM classifier. Experimental results show that the proposed system for recognition of Kufic script performs better than the previous recognition systems for Arabic text. [ABSTRACT FROM AUTHOR]

• This is the first corpus-based study on antonym sequence in Modern Standard Arabic. • Antonym sequence rules across languages are replicable with Modern Standard Arabic. • Positivity and morphology figure more prominently in Modern Standard Arabic. • Some Arabic cases show intervention and overlap of antonym sequence rules. • Antonym sequence rules are occasionally overruled for contextual reasons. The present study investigates the rules which govern antonym sequences in Modern Standard Arabic (MSA) by means of triangulation, i.e., a mixed-method design of quantification and qualification, whereby corpus tools are utilized to identify and examine antonym sequences in the Arabic corpus arTenTen18 integrated in Sketch Engine online interface. Ranging from qualitative to quantitative analyses, several studies took the order of antonym pairs in various languages as their focus, but no previous study examined this linguistic phenomenon in MSA using corpus-based methods. To give a panoramic account and conduct a rigorous analysis of antonym sequence in MSA, the current study combines both quantitative and qualitative approaches based upon studies across English, Serbian, Chinese, Persian, and Qur'anic Arabic, respectively. The study concludes that all the previously investigated antonym sequence rules drawn from English, Serbian, Chinese, Persian, and Qur'anic Arabic are replicable with an MSA corpus. Moreover, the rules controlling the co-occurrences of antonym pairs were also found to occasionally overlap. Contextual factors were found to play a crucial role in determining which rules are operative and which are solidly overruled. [ABSTRACT FROM AUTHOR]

The advent of online social networks (OSNs) has catalyzed the formation of novel learning communities. Identifying experts within OSNs has become a critical component for facilitating knowledge exchange and enhancing self-awareness, particularly in contexts such as rumor verification processes. Research efforts aimed at locating authorities in OSNs are scant, largely due to the scarcity of annotated datasets. This work represents a contribution to the domain of unsupervised learning to address the challenge of authorities' identification in Twitter. We have employed advanced natural language processing technique to transfer knowledge concerning topics in the Arabic language and to discern the semantic connections among candidates within Twitter in zero-shot learning. We take advantage of the Single-labeled Arabic News Articles Dataset (SANAD) to perform the process of extracting domain features and applying these features in finding authorities using the Authority Finding in Arabic Twitter (AuFIN) dataset. Our evaluation assessed the extent of extracted topical features transferred and the efficacy of authorities' retrieval in comparison to the latest unsupervised models in this domain. Our approach successfully extracted and integrated the limited available topical semantic features of the language into the representation of candidates. The findings indicate that our hybrid model surpasses those that rely solely on lexical features of language and network topology, as well as other contemporary approaches to topic-specific expert finding. [ABSTRACT FROM AUTHOR]

Question answering (QA) is one of the essential fields in information retrieval where specific answers are provided instead of large documents. The relations among questions and answers are determined using natural language processing techniques while clustering algorithms can be helpful in improving the effectiveness of result retrieval by reducing the amount of required comparisons for a specific question or answer. In this work, we introduce a clustering-based approach for a QA system. This approach groups related questions into clusters using different clustering algorithms, specifies the appropriate answer using similarity methods between the answers and the generated clusters, and then assigns answers to their most related questions. Different clustering algorithms, such as k-means, spherical k-means, single-linkage hierarchical clustering (SLHA), unweighted pair group method with arithmetic mean (UPGMA), expectation–maximization (EM), and clustering Arabic documents based on bond energy (CADBE), are tested. The effectiveness of a clustering algorithm is investigated with respect to certain factors, including number of clusters, text representation, similarity measure between answers and clusters, and similarity measure between answers and questions in a selected cluster. In addition, a comprehensive ranking system is introduced to evaluate the performance of clustering algorithms. Evaluation is performed using the Dataset of Arabic Why Question Answering System (DAWQAS) and the Multilingual Question Answering (MLQA) dataset. Results show that CADBE achieves the highest accuracy and the first rank, followed by SLHA and UPGMA, while spherical k-means has the lowest rank. The performance of clustering algorithms for MLQA dataset is affected by its characteristics, such as short questions, long and varied answers, and diverse subject domains. Unigram and bigram intersection measures perform well in most cases. Term frequency inverse document frequency representation outperforms word embedding in DAWQAS. Overall, the experiments provide insights into the performance of clustering algorithms in QA systems. • A clustering-based QA system groups related questions, selects answer via similarity. • Assigning Answers to Related Questions Using Various Similarity Methods. • Exploring certain factors to investigate effectiveness of clustering algorithm. • A comprehensive ranking system evaluates the performance of clustering algorithms. • CADBE achieves highest accuracy, then SLHA, UPGMA. Spherical k-means ranks lowest. [ABSTRACT FROM AUTHOR]

This study explores X's (formerly Twitter's) capacity to serve as a real-time barometer of public sentiment, contextualized within the transformative reforms of Saudi Arabia during 2016–2017. The objective was to decipher the populace's response to these significant national changes by analyzing approximately 200 million tweets in native Arabic dialects, thereby aiming for an authentic portrayal of local sentiment. Our methodology entailed a dual-phase analysis: initial tweet examination to discern prevalent social behaviors, followed by stance detection to classify tweets according to their support, neutrality, or opposition to the divisive issues at hand. For sentiment extraction, we employed a sophisticated feature vector, integrating the k most frequent words and stems. A comprehensive evaluation of various classifiers was conducted, including Support Vector Machine and several variants of K -nearest neighbors (K -NN), with a particular emphasis on their applicability to our dataset. Notably, the 9-NN classifier, and more specifically, the weighted K -NN approach, demonstrated remarkable performance, achieving an F -score of 72.45%. These insights not only shed light on the public's reception to the Saudi reforms but also position Twitter as a viable, real-time alternative to traditional survey methods for capturing the nuances of public opinion, thereby offering valuable perspectives for policy formulation. [ABSTRACT FROM AUTHOR]

With the large amount of unstructured data being broadcasted every day, building powerful methods enabling information retrieval and extraction becomes necessary. Unfortunately, named entity recognition is a difficult classification task to classify data into predefined labels, which is further challenged by the Arabic language's particular characteristics and complex nature. This work trains six BERT-based models (Bidirectional Encoder Representations from Transformers) and uses a BiLSTM-CRF architecture for the NER task on dialectal Arabic. Our fine-tuning approach yields new state-of-the-art results on publicly available dialectal Arabic social media datasets. [ABSTRACT FROM AUTHOR]

Fake news detection for Arabic news has drawn some attention recently. However, the number of such studies are limited due to the lack of datasets that can be used to perform them. Clickbait detection is typically linked to fake news detection as clickbaits are effective in spreading fake news. The lack of dataset in the Arabic language to study clickbait detection models is also evident. This paper presents a dataset of Arabic clickbait news for the first time. The purpose of this dataset is to enable the automatic classification of news headlines as "Clickbait" or "Not Clickbait" using a machine learning model. More than 3000 news records are sampled from five months of tweets for 24 Jordanian news publishers. All sampled news records are labeled by three annotators and that resulted in 18% clickbait news records. The annotator unanimously agreed on the class of about 81% of the labeled news records. To showcase the usability of the resulting dataset in machine learning, Logistic Regression, Support Vector Machine, Random Forrest, Naïve Bayes, Stochastic Gradient Descent, Nearest Neighbor, and Decision Tree are applied to this dataset. These models produced Macro F1-Score value up to 0.81 indicating that the automatic detection of clickbait news headlines using machine learning is feasible. [ABSTRACT FROM AUTHOR]

This research aims to identify the geographic origin of Arabic-speaking social media users by analyzing textual data they produce and share. The paper presents an approach to infer users' region (i.e country) of origin through identification of the dialect they use in their written interactions. An Integrated Dataset for Arabic Dialect Detection (IADD) is proposed and used to train multiple classifiers which succeed in identifying the users' region and country of origin with an accuracy of 0.89 and 0.93, respectively. [ABSTRACT FROM AUTHOR]

ATLASLang is a machine translation system from Arabic text language into Arabic sign language (ArSL). The first version of the system (Brour and Benabbou, 2019) is based on two approaches: rule-based Interlingua and example-based approaches. It is a classical machine translation system, its limitations are the linguistic knowledge necessary to develop the rules, also many rules and exceptions required. In the last few years, a neural machine translation has achieved notable results, and several well-known companies including Google (Wu et al., 2016) and Systran (Crego et al., 2016) have starting to exploit it. In this paper, we present a new version of ATLASLang that is implemented using a feed-forward back-propagation Artificial Neural Network. In this version, we have started to translate simple sentences composed from a limited number of words since generally communication with deaf person uses short sentences. We have used the same ATLASLang MTS database of signs and we have utilized morphological characteristics to derive the maximum information from each word in the input of the neural translation system. The two versions of the system are compared using n-gram BLEU score (Papineni et al., 2002), and the results demonstrate that neuron network approaches outperform the other classical approaches. [ABSTRACT FROM AUTHOR]

Steganography is the process of hiding data within multimedia communication between parties by embedding the secret data inside a carrier file to be protected during its transmission. This paper addresses the problem of hiding information within the Arabic text as it is currently a challenging research area. Two methods are presented for text steganography: utilizing pseudo-spaces alone and combining them with Kashida (extension character). Experimental results have shown that the proposed algorithms achieved high capacity and security ratio as compared to state-of-the-art steganography methods presented for the Arabic language. The proposed techniques are of great benefit and can be further used for languages similar to Arabic, such as Urdu and Persian, as well as opening the direction of text-stego research for other languages of the world. [ABSTRACT FROM AUTHOR]

As part of a research project in computer science on the realization of a cryptographic system using Arabic poems as secret keys (Azizi, 2012), we defined the formal structure of the rhythms of Arabic words. These rhythms, as they were studied by Al Khalil Al Farahidi (▪), are the sequences of movement and silence, which take into account only the pronounced letters. The metric prosody of the Arabic language was designed by its founders in a fundamentally quantitative way. We will demonstrate in this paper that the rhythms of Arabic words are variants of Fibonacci words. [ABSTRACT FROM AUTHOR]

In linguistic steganography, languages' features are employed to hide information. The Arabic language has a rich set of features which have not been utilized until now in this area. In particular, Arabic calligraphy contains multiple fonts and multiple shapes of Arabic alphabet letters. In this paper, a framework that uses Arabic calligraphy to hide information is proposed. The phases of the framework are preparation phase, embedding phase and extraction phase. The embedding phase uses string matching to generate stego text and accompanying letter shapes according to a secret message. The framework also includes corpus creation and a modification of the Aho-Corasick string-matching algorithm. The Arabic font Naskh was used as a case study. A set of Arabic poetry and proverbs were used as a dataset. The framework was evaluated on capacity and security. Because the visual difference between the cover and the stego-cover must be unnoticeable to the human in any stego-system, the security in this framework is satisfying due there is no cover used. The cover represents the secret message itself and it provides high capacity to hide data also. The evaluation showed the potential of using the multiple shapes of Arabic letters to satisfy steganography requirements. [ABSTRACT FROM AUTHOR]

In this paper we propose a new approach to analyze sentiments for the Arabic language. To overcome the scarcity and size limitation of the required Arabic language resources for training and analysis tasks, we built new lexical resources using different approaches. We have also integrated the morphological notion by creating both stemmed and lemmatized versions of word lexicons. Thereafter, the generated resources were used in the construction of a supervised model from a set of features considering the word negation context. Similarly, we have semantically segmented the lexicon in order to reduce the model vectors size and consequently improve the execution time. [ABSTRACT FROM AUTHOR]

Multi-label classification assigns multiple labels to each document concurrently. Many real-world classification problems tend to employ high-dimensional label spaces, which can be naturally structured in a hierarchy. In this type of problem, each instance may belong to multiple labels and labels are organized in a hierarchical structure. It presents a more complex problem than flat classification, given that the classification algorithm has to take into account hierarchical relationships between labels and be able to predict multiple labels for the same instance. Few studies have investigated multi-label text classification for the Arabic language. Most of these studies have focused mainly on flat classification and have neglected the hierarchical structure. Therefore, this paper explores the hierarchical multi-label classification in the context of the Arabic language. It proposes a hierarchical multi-label Arabic text classification (HMATC) model with a machine learning approach. The impact of feature selection methods and feature set dimensions on classification performance are also investigated. In addition, the Hierarchy Of Multilabel ClassifiER (HOMER) algorithm is optimized via examination of different sets of multi-label classifiers, clustering algorithms and different numbers of clusters to improve the hierarchical classification. Moreover, this study contributes to existing research by introducing a hierarchical multi-label Arabic dataset in an appropriate format for hierarchical classification and making it publicly available. The results reveal that the proposed model outperforms all models considered in the experiments in terms of the computational cost, which consumed less cost (2 h) compared with other evaluated models. In addition, it shows a significant improvement compared with the state-of-the-art model (Fatwa model) in terms of Hamming loss (0.004), hierarchical loss (1.723), multi-label accuracy (0.758), subset accuracy (0.292), micro-averaged precision (0.879), micro-averaged recall (0.828), and micro-averaged F-measure (0.853). [ABSTRACT FROM AUTHOR]

There is little pragmatic research on the pluricentric nature of the Arabic language. The current study highlights this nature by adopting a variational pragmatic approach to examine the complaint behavior of speakers of two Arabic dialects: Alexandrian Arabic versus Najdi Arabic. Data were collected from 120 undergraduates through roleplays and coded using an adapted version of Trosborg's (1995) coding scheme for complaints and internal modifiers. The findings showed that both groups preferred using directive acts followed by expressions of disapproval and blame in their complaints. However, the use of hints and accusations was minimized. The study found that Alexandrians employ more direct complaint strategies and exhibit less influence of the social variables of gender, social distance, and social dominance than Najdis. Additionally, both groups show concern for a complainee's negative face but to varying degrees. The results are interpreted in the context of existing literature, politeness theory, and relevant cultural models. • Alexandrians and Najdis prefer to use directive acts to complain. • Hints and accusations are used to the minimum by Alexandrians and Najdis. • Alexandrians exhibit a more direct/pragmatic approach to complaints than Najdis. • Alexandrians and Najdis are highly influenced by social distance and dominance. • The current study calls for revising several politeness and cultural models. [ABSTRACT FROM AUTHOR]

Although a huge amount of data has been provided on the web to meet different needs of users, it has been more difficult to quickly make satisfactory choices. Therefore, recommender systems have been introduced for offering relevant data and providing users with personalized recommendations. Traditional collaborative based methods modelled efficiently users and items from their rating patterns. By cons, they suffered from the data sparsity problem. As a solution, text user-generated data (e.g., reviews) were useful to provide rich semantic information of user preferences and item features. Seeing that opinions and ratings were complementary, we proposed to fuse them and learn more accurate representations of users and items for recommendation. Unlike Arabic which has received less attention, the majority of methods offered have been focused on English language. It has been represented a fundamental issue in Natural Language Processing (NLP) and artificial intelligence. In this paper, a neural collaborative embedding- and filtering-based approach was proposed for Arabic textual documents recommendation. Due to the recent advances on neural networks, our idea was to couple deep features learning- and deep interaction modelling- based encoders. Indeed, two branches of recurrent neural networks were combined to extract latent features of users and items. Using the generated representations as inputs, deep user-item interactions were captured by stacking multiple layers. This method has proved its efficiency to simultaneously capture the global context and the word frequency information, learn thereafter the obtained high-level latent features and produce final ratings. Extensive experiments on a standard Arabic dataset showed the superiority of our model than state-of-the art methods. [ABSTRACT FROM AUTHOR]

The exponential growth of online textual data triggered the crucial need for an effective and powerful tool that automatically provides the desired content in a summarized form while preserving core information. In this paper, we propose an automatic, generic, and extractive Arabic single document summarizing method aiming at producing a sufficiently informative summary. The proposed extractive method evaluates each sentence based on a combination of statistical and semantic features in which a novel formulation is used taking into account sentence importance, coverage and diversity. Further, two summarizing techniques including score-based and supervised machine learning were employed to produce the summary and then assist leveraging the designed features. We demonstrate the effectiveness of the proposed method through a set of experiments under EASC corpus using ROUGE measure. Compared to some existing related work, the experimental evaluation shows the strength of the proposed method in terms of precision, recall, and F-score performance metrics. [ABSTRACT FROM AUTHOR]

The automatic processing of natural languages knew a remarkable growth in terms of techniques, methods and the variety of its potential applications. For instance, to automatize the processing of the Arabic language with its particular features, we selected the holy Quran as a study case. The latter has represented a challenge for the artificial intelligence. The main objective of the present paper is to design the semantic search engine for the text of the Quran using Quranic ontology. To determine the semantic fields of the words of the holy Quran, the Quranic ontology was developed, that presents the meaning of words and their relations. This method is used for each concept in order to enrich the query. [ABSTRACT FROM AUTHOR]

Modeling a natural language aims to build systems that are able to reproduce, correct and more or less predict the structure of a given language; moreover it summarizes the general knowledge related to that language. In this context, we have suggested a new language model which efficiently estimate the n-gram language model. This new model has made it possible to remedy the shortcomings of the n-gram language model. By using an automatic spelling error application, the evaluation of this model has been done, and after the comparison of the results obtained; we see that this new model give quick and precise results. These results show that this new proposed model better models the succession of words within sentences, and can therefore replace the n-gram model in Natural Language Processing (NLP) applications. [ABSTRACT FROM AUTHOR]

Automatic captioning of images no only enrich multimedia content with descriptive features, but also helps in detecting patterns, trends, and events of interest. Particularly, Arabic Image Caption Generation is a very challenging topic in the machine learning field. This paper presents, AraCap, a hybrid object-based, attention-enriched image captioning architecture, with a focus on Arabic language. Three models are demonstrated, all of them are implemented and trained on COCO and Flickr30k datasets, and then tested by building an Arabic version of a subset of COCO dataset. The first model is an object-based captioner that can handle one or multiple detected objects. The second is a combined pipeline that uses both object detector and attention-based captioning; while the third one is based on a pure soft attention mechanism. The models are evaluated using multi-lingual semantic sentence similarity techniques to assess the generated captions accuracy against the actual ground truth captions. Results show that similarity scores for Arabic generated captions from all three proposed models outperformed the basic captioning technique. [ABSTRACT FROM AUTHOR]

Topic modeling is an unsupervised machine learning technique for finding abstract topics in a large collection of documents. It helps in organizing, understanding and summarizing large collections of textual information and discovering the latent topics that vary among documents in a given corpus. Latent Dirichlet allocation (LDA) and Non-Negative Matrix Factorization (NMF) are two of the most popular topic modeling techniques. LDA uses a probabilistic approach whereas NMF uses matrix factorization approach, however, new techniques that are based on BERT for topic modeling do exist. In this paper, we aim to experiment with BERTopic using different Pre-Trained Arabic Language Models as embeddings, and compare its results against LDA and NMF techniques. We used Normalized Pointwise Mutual Information (NPMI) measure to evaluate the results of topic modeling techniques. The overall results generated by BERTopic showed better results compared to NMF and LDA. [ABSTRACT FROM AUTHOR]

There are more than 80 million students in the Arab world. Students take tests and examinations as an intrinsic part of their educational journey to assess the quality of learning and understanding of the examined material. Exams require an immense amount of resources to be conducted, but even a bigger amount of resources to be scored. Scoring of exams and papers takes up enormous time and effort from teachers all over the Arab world. This time and effort can be better utilized in other teaching activities that would elevate the quality of learning in the region. For that reason, the work presented in this paper investigates multiple supervised learning approaches for the domain of Arabic short answer scoring. Moreover, due to the scarcity of Arabic datasets, and to our knowledge, there is only one publicly available dataset for this specific task(AR-ASAG). Thus, we introduce a new publicly available dataset AraScore-Dataset. The model proposed is evaluated on 3 different datasets: 1-AraScore-Dataset 2-AR-ASAG 3-Two translated answer sets taken from the Hewlett Foundation SAS dataset. Based on the investigation, the most effective approach for Arabic automated short answer scoring is proposed, and the performance of using our dataset is compared to other publicly available datasets. The aim is to scale efficient and unbiased content scoring applications across different Arabic educational domains. The results showed that our newly developed response-based system and AraScore-Dataset have achieved state-of-the-art performance on Arabic Automated Short Answer Scoring. [ABSTRACT FROM AUTHOR]

Along with the COVID-19 pandemic, an "infodemic" of false and misleading information has emerged and has complicated the COVID-19 response efforts. Social networking sites such as Facebook and Twitter have contributed largely to the spread of rumors, conspiracy theories, hate, xenophobia, racism, and prejudice. To combat the spread of fake news, researchers around the world have and are still making considerable efforts to build and share COVID-19 related research articles, models, and datasets. This paper releases "AraCOVID19-MFH"1 a manually annotated multi-label Arabic COVID-19 fake news and hate speech detection dataset. Our dataset contains 10,828 Arabic tweets annotated with 10 different labels. The labels have been designed to consider some aspects relevant to the fact-checking task, such as the tweet's check worthiness, positivity/negativity, and factuality. To confirm our annotated dataset's practical utility, we used it to train and evaluate several classification models and reported the obtained results. Though the dataset is mainly designed for fake news detection, it can also be used for hate speech detection, opinion/news classification, dialect identification, and many other tasks. [ABSTRACT FROM AUTHOR]

As a key player in today's world, online social networks are emerging, providing a platform for expression and content distribution. This technology enables users to communicate easily with each other and share their data instantly. However, the internet isn't generally protected; it can be a source for abusive and harmful content and causing harm to others. There is a great need for approaches and strategies to solve these issues due to the negative effect of abusive language and cyberbullying. Arabic text is known for its challenges, complexity, and scarcity of its resources. Many languages have made many efforts to find automated solutions for detecting abusive language and cyberbullying, but not much for the Arabic language. This work analyzes 27 studies on automatic Arabic abusive language and cyberbullying and its related detection approaches. The goal of this paper is to review the findings of the previous studies about cyberbullying and abusive detection in Arabic content on online social networks and help researcher in the future to develop automatic detection systems that are effective and realistic. [ABSTRACT FROM AUTHOR]

Some users try to post false reviews to promote or to devalue other's products and services. This action is known as deceptive opinions spam, where spammers try to gain or to profit from posting untruthful reviews. Therefore, we conducted this work to develop and to implement new semantic features to improve the Arabic deception detection. These features were inspired from the study of discourse parse and the rhetoric relations in Arabic. Looking to the importance of the phrase unit in the Arabic language and the grammatical studies, we have analyzed and selected the most used unit markers and relations to calculate the proposed features. These last were used basically to represent the reviews texts in the classification phase. Thus, the most accurate classification technique used in this area which has been proven by several previous works is the Support Vector Machine classifier (SVM). But there is always a lack concerning the Arabic annotated resources specially for deception detection area as it is considered new research area. Therefore, we used the semi supervised SVM to overcome this problem by using the unlabeled data. [ABSTRACT FROM AUTHOR]

Arabic is recognised as the 4th most used language of the Internet. Arabic has three main varieties: (1) classical Arabic (CA), (2) Modern Standard Arabic (MSA), (3) Arabic Dialect (AD). MSA and AD could be written either in Arabic or in Roman script (Arabizi), which corresponds to Arabic written with Latin letters, numerals and punctuation. Due to the complexity of this language and the number of corresponding challenges for NLP, many surveys have been conducted, in order to synthesise the work done on Arabic. However these surveys principally focus on two varieties of Arabic (MSA and AD, written in Arabic letters only), they are slightly old (no such survey since 2015) and therefore do not cover recent resources and tools. To bridge the gap, we propose a survey focusing on 90 recent research papers (74% of which were published after 2015). Our study presents and classifies the work done on the three varieties of Arabic, by concentrating on both Arabic and Arabizi, and associates each work to its publicly available resources whenever available. [ABSTRACT FROM AUTHOR]

Natural Language Processing (NLP) has been a growing area of research in computer and cognitive sciences, using experimental approaches. Morphology and syntax play specifically a vital role in the correct interpretation of a sentence. In this paper, we will present a syntactic error correction system based on the automatic generation of correct sentences in Arabic. First, we extract the words from the considered sentence and we then generate all the possible sentences that are syntactically correct; based on a logical description of the rules of Arabic grammar in the ontology. We will afterwards compare the original sentence with the generated sentences to detect any eventual errors followed by the correction phase. In case the system has not found a sentence that looks similar to the original sentence, the correct alternative sentences are automatically offered. The use of the Arabic syntactic corrector can increase productivity and improve the quality of the text for anyone who writes in the Arabic language. Successful tests have been performed using a set of Arabic sentences. The implemented system achieved a precision rate of about 92% and a recall rate of about 84%. By observing the achieved results, it is concluded that our approach is promising. [ABSTRACT FROM AUTHOR]