Arabic Dataset Research Articles

Leveraging Social Media and Deep Learning for Sentiment Analysis for Smart Governance: A Case Study of Public Reactions to Educational Reforms in Saudi Arabia

The Saudi government’s educational reforms aim to align the system with market needs and promote economic opportunities. However, a lack of credible data makes assessing public sentiment towards these reforms challenging. This research develops a sentiment analysis application to analyze public emotional reactions to educational reforms in Saudi Arabia using AraBERT, an Arabic language model. We constructed a unique Arabic dataset of 216,858 tweets related to the reforms, with 2000 manually labeled for public sentiment. To establish a robust evaluation framework, we employed random forests, support vector machines, and logistic regression as baseline models alongside AraBERT. We also compared the fine-tuned AraBERT Sentiment Classification model with CAMeLBERT, MARBERT, and LLM (GPT) models. The fine-tuned AraBERT model had an F1 score of 0.89, which was above the baseline models by 5% and demonstrated a 4% improvement compared to other pre-trained transformer models applied to this task. This highlights the advantage of transformer models specifically trained for the target language and domain (Arabic). Arabic-specific sentiment analysis models outperform multilingual models for this task. Overall, this study demonstrates the effectiveness of AraBERT in analyzing Arabic sentiment on social media. This approach has the potential to inform educational reform evaluation in Saudi Arabia and potentially other Arabic-speaking regions.

A Transfer Learning Approach for Arabic Image Captions

Background: Arabic image captioning (AIC) is the automatic generation of text descriptions in the Arabic language for images. Applies a transfer learning approach in deep learning to enhance computer vision and natural language processing. There are many datasets in English reverse other languages. Instead of, the Arabs researchers unanimously agreed that there is a lack of Arabic databases available in this field. Objective: This paper presents the improvement and processing of the available Arabic textual database using Google spreadsheets for translation and creation of AR. Flicker8k2023 dataset is an extension of the Arabic Flicker8k dataset available, it was uploaded to GitHub and made public for researches. Methods: An efficient model proposed using deep learning techniques by including two pre-training models (VGG16 and VGG19), to extract features from the images and build (LSTM and GRU) models to process textual prediction sequence. In addition to the effect of pre-processing the text in Arabic. Results: The adopted model outperforms better compared to the previous study in BLEU-1 from 33 to 40. Conclusions: This paper concluded that the biggest problem is the database available in the Arabic language. This paper has worked to increase the size of the text database from 24,276 to 32,364 thousand captions, where each image contains 4 captions.

Amina: an Arabic multi-purpose integral news articles dataset

AbstractElectronic newspapers are one of the most common sources of Modern Standard Arabic. Existing datasets of Arabic news articles typically provide a title, body, and single label. Ignoring important features, like the article author, image, tags, and publication date, can degrade the efficacy of classification models. In this paper, we propose the Arabic multi-purpose integral news articles (AMINA) dataset. AMINA is a large-scale Arabic news corpus with over 1,850,000 articles collected from 9 Arabic newspapers from different countries. It includes all the article features: title, tags, publication date and time, location, author, article image and its caption, and the number of visits. To test the efficacy of the proposed dataset, three tasks were developed and validated: article textual content (classification and generation) and article image classification. For content classification, we experimented the performance of several state-of-the-art Arabic NLP models including AraBERT and CAMeL-BERT, etc. For content generation, the reformer architecture is adopted as a character text generation model. For image classification applied on Al-Sharq and Youm7 news portals, we have compared the performance of 10 pre-trained models including ConvNeXt, MaxViT, ResNet18, etc. The overall study verifies the significance and contribution of our newly introduced Arabic articles dataset. The AMINA dataset has been released at https://huggingface.co/datasets/MohamedZayton/AMINA.

Spam Review Detection Using Weighted Swarm Support Vector Machines and Pre-Trained Word Embedding for Multiple Languages

Before making a purchase, many find internet reviews to be a vital source of information. On top of that, companies may learn a lot about their products and services via these reviews. Having faith in these reviews was especially important during the COVID-19 pandemic, when many stayed inside and read reviews at a dizzying pace. The pandemic altered the atmosphere and people's preferences in addition to increasing the number of evaluations. Spam reviewers keep an eye on these changes and try to improve their sneaky techniques. In order to deceive customers or harm competitors, reviews that are deemed spam may include inaccurate, misleading, or dishonest information. Consequently, this work introduces a WSVM plus an HHO to identify spam reviews. The HHO is similar to an algorithm in that it optimises hyperparameters and uses feature weights. Using English, Spanish, and Arabic language corpora as datasets, the multilingual difficulty in spam reviews has been tackled. Ngram-3, TFIDF, whereas One-hot encoding are three methods for representing words, while pre-trained word incorporation (BERT) is another one that has been used. Each of the four such studies has shed light on and provided a solution to a different facet. From start to finish, the proposed technique beat rival cutting-edge algorithms in every test. For the Multi dataset, the WSVM-HHO achieved a success rate of 84.270 percent; for the English information set, 89.565 percent; for the Spanish information set, 71.913 percent; and for the Arabic dataset, 88.565 percent. Furthermore, we have extensively researched the review environment before to and during the COVID-19 event. To further enhance detection performance, it has been designed to merge its existing textual attributes with statistical information to build a new dataset.

Hybrid CNN-BiLSTM architecture with multiple attention mechanisms to enhance speech emotion recognition

During recent years, the concept of attention in deep learning has been increasingly used to boost the performance of Speech Emotion Recognition (SER) models. However, these models for SER exhibit shortcomings in jointly emphasizing the time-frequency and dynamic sequential variations, often under-utilizing the rich contextual emotion-related information. We propose a hybrid deep learning model for SER using Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory Networks (BiLSTM) with multiple attention mechanisms. Our model utilizes features from the speech waveform viz. Mel spectrograms and Mel Frequency Cepstral Coefficients (MFCC), along with their time derivatives as input to the CNN and BiLSTM modules, respectively. A Time–Frequency Attention (TFA) mechanism, optimally incorporated into CNN, helps to selectively focus on emotion-related energy–time–frequency variations in Mel spectrograms. Attention-based BiLSTM uses MFCC and its time derivatives to identify the positional information of emotion for addressing the dynamic sequential variations. Finally, we fuse the attention-learned features from the CNN and BiLSTM modules and feed them to a Deep Neural Network (DNN) for SER. The experiments were carried out using three different datasets: Emo-DB and IEMOCAP, which are public datasets, and Amritaemo_Arabic; a private dataset. The hybrid model exhibited superior performance on both the public and private datasets, generating an average SER accuracy of 94.62%, 67.85%, and 95.80% with Emo-DB, IEMOCAP, and Amritaemo_Arabic datasets, respectively, effectively outperforming several state-of-the-art models.

Deciphering Arabic question: a dedicated survey on Arabic question analysis methods, challenges, limitations and future pathways

This survey reviews different research on question analysis, including other comparative studies of question analysis approaches and an evaluation of the questions by different NLP techniques that are used in question interpretation and categorization. Among these key findings noted includes the assessment of deep learning models such as M-BiGRU-CNN and M-TF-IDF, which come with high precision and accuracy when applied with the effectiveness of use in dealing with the complexities involved in a language. Some of the most mature machine learning algorithms, for example, SVM or logistic regression, remain powerful models, especially on the classification task, meaning that the latter continues to be relevant. This study further underlines the applicability of rule-based or hybrid methodologies in certain linguistic situations, and it must be said that custom design solutions are required. We could recommend, on this basis, directing future work towards the integration of these hybrid systems and towards the definition of more general methodologies of evaluation that are in line with the constant evolution of NLP technologies. It revealed that the underlying challenges and barriers in the domain are very complex syntactic and dialectic variations, unavailability of software tools, very critical standardization in Arabic datasets, benchmark creation, handling of translated data, and the integration of Large Language Models (LLMs). The paper discusses the lack of identity and processing of such structures through online systems for comparison. This comprehensive review highlights not only the diversified potential for the capabilities of NLP techniques in refining question analysis but also the potential way of great promises for further enhancements and improvements in this progressive domain.

Chimp Optimization Algorithm with Deep Learning-Driven Fine-grained Emotion Recognition in Arabic Corpus

Recently, emotion analysis and classification of tweets have become a crucial area of research. The Arabic language had experienced difficulties with emotion classification on Twitter(X), needing preprocessing more than other languages. Emotion detection is a major challenge in Natural Language Processing (NLP), which allows machines to ascertain the emotions expressed in the text. The task includes recognizing and identifying human feelings such as fear, anger, sadness, and joy. The discovered sentiments and feelings expressed in tweets have gained much recognition in recent years. The Arab region has played a substantial role in international politics and the global economy needs to scrutinize the emotions and sentiments in the Arabic language. Lexicon-based and machine-learning techniques are two common models that address the problems of emotion classification. This study introduces a Chimp Optimization Algorithm with a Deep Learning-Driven Arabic Fine-grained Emotion Recognition (COADL-AFER) technique. The presented COADL-AFER technique mainly aims to detect several emotions in Arabic tweets. In addition to its academic significance, the COADL-AFER technique has practical applications in various fields, including enhancing applications of E-learning, aiding psychologists in recognising terrorist performance, improving product quality, and enhancing customer service. The COADL-AFER technique applies the long short-term memory (LSTM) model for emotion detection. Finally, the hyperparameter selection of the LSTM method can be accomplished by COA. The experimental validation of the COADL-AFER system, a crucial step in our research, is verified utilizing the Arabic tweets dataset. The simulation results stated the betterment of the COADL-AFER technique, further reinforcing the reliability of our research.

2D Spectrogram analysis using vision transformer to detect mispronounced Arabic utterances for children

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.

MSAPersonality: a modern standard Arabic dataset for personality recognition

Automatic personality recognition is a task that attempts to automatically infer personality traits from a variety of data sources, including Text. Our words, whether spoken or written, reveal a lot about who we are. As people speak different languages, each with its own set of characteristics and level of complexity, identifying their personalities automatically might be language-dependent. This task requires an annotated text corpus with personality traits. However, the lack of corpora for languages other than English makes the task extremely challenging. We concentrated our efforts in this paper on the Arabic language in particular because it is understudied and lacks a corpus, despite being one of the most widely spoken languages in the world. Our primary goal was constructing our “MSAPersonality” dataset, which consists of 267 texts in modern standard Arabic that have been annotated with the Big Five personality traits. To evaluate the dataset and its potential for classification and regression, we used text preprocessing techniques, feature extraction, and machine learning algorithms. We obtained promising experimental results. Therefore, further research into predicting personality from Arabic text can be conducted.

ArabBert-LSTM: improving Arabic sentiment analysis based on transformer model and Long Short-Term Memory.

Sentiment analysis also referred to as opinion mining, plays a significant role in automating the identification of negative, positive, or neutral sentiments expressed in textual data. The proliferation of social networks, review sites, and blogs has rendered these platforms valuable resources for mining opinions. Sentiment analysis finds applications in various domains and languages, including English and Arabic. However, Arabic presents unique challenges due to its complex morphology characterized by inflectional and derivation patterns. To effectively analyze sentiment in Arabic text, sentiment analysis techniques must account for this intricacy. This paper proposes a model designed using the transformer model and deep learning (DL) techniques. The word embedding is represented by Transformer-based Model for Arabic Language Understanding (ArabBert), and then passed to the AraBERT model. The output of AraBERT is subsequently fed into a Long Short-Term Memory (LSTM) model, followed by feedforward neural networks and an output layer. AraBERT is used to capture rich contextual information and LSTM to enhance sequence modeling and retain long-term dependencies within the text data. We compared the proposed model with machine learning (ML) algorithms and DL algorithms, as well as different vectorization techniques: term frequency-inverse document frequency (TF-IDF), ArabBert, Continuous Bag-of-Words (CBOW), and skipGrams using four Arabic benchmark datasets. Through extensive experimentation and evaluation of Arabic sentiment analysis datasets, we showcase the effectiveness of our approach. The results underscore significant improvements in sentiment analysis accuracy, highlighting the potential of leveraging transformer models for Arabic Sentiment Analysis. The outcomes of this research contribute to advancing Arabic sentiment analysis, enabling more accurate and reliable sentiment analysis in Arabic text. The findings reveal that the proposed framework exhibits exceptional performance in sentiment classification, achieving an impressive accuracy rate of over 97%.

Arabic Chatbots Challenges and Solutions: A Systematic Literature Review

Since the beginning of Natural Language Processing, researchers have been highly interested in the idea of equipping machines with the ability to think, understand, and communicate in a human-like manner. However, despite the significant progress in this field, particularly in English, Arabic research remains in its early stages of development. This study presents a Systematic Literature Review on challenges faced in Arabic chatbot development and the proposed solutions. Utilizing the search terms "ARABIC," "CHATBOT," "CHALLENGES," and "SOLUTION," (including synonyms) we systematically surveyed studies published between 2000 and 2023 from Scopus, Science Direct, Web of Science, PubMed, SpringerLink, IEEE Xplore, ACM, Ebesco, and ICI. Moreover, besides Google Scholar and ResearchGate, we employed a manual snowballing technique to discover supplementary relevant research by examining the references of all chosen primary studies. The included studies were assessed for eligibility based on the quality assessment checklist we developed. Out of 3,891 studies, only 64 were deemed eligible. Many challenges were identified, including the scarcity of well-structured datasets. To overcome this (n=35) studies manually collected and preprocessed data. Additionally, Arabic language complexity led (n=53) researchers to adopt pre-scripted rules approaches, followed by generative approaches (n=10), and hybrid approach (n=1). Furthermore, (n=27) studies employed human-based evaluation metrics to assess the chatbot performance. while, (n=11) studies haven’t used any metrics. Based on conducted research, a critical research priority is providing Arabic with high-quality resources, such as an Arabic dataset that includes dialectal variations and incorporates empathy, lexicon corpora, and also a word normalization library. These resources will enable the chatbot to interact more naturally and humanely. Additionally, hybrid approaches have shown promising results, particularly in low-resource languages, such as Arabic. Therefore, more focus should be dedicated into implementing hybrid approaches in chatbot development. Furthermore, evaluating the chatbot performance is still an open domain for further research and contribution, highlighting the need for innovative standardized evaluation methods.

Rumor gatekeepers: Unsupervised ranking of Arabic twitter authorities for information verification

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.

Neural multi-task learning for end-to-end Arabic aspect-based sentiment analysis

Most existing aspect-based sentiment analysis (ABSA) methods perform the tasks of aspect extraction and sentiment classification independently, assuming that the aspect terms are already determined when handling the aspect sentiment classification task. However, such settings are neither practical nor appropriate in real-life applications, as aspects must be extracted prior to sentiment classification. This study aims to overcome this shortcoming by jointly identifying aspect terms and the corresponding sentiments using a multi-task learning approach based on a unified tagging scheme. The proposed model uses the Bidirectional Encoder Representations from Transformers (BERT) model to produce the input representations, followed by a Bidirectional Gated Recurrent Unit (BiGRU) layer for further contextual and semantic coding. An attention layer is added on top of BiGRU to force the model to focus on the important parts of the sentence. Finally, a Conditional Random Fields (CRF) layer is used to handle inter-label dependencies. Experiments conducted on a reference Arabic hotel dataset show that the proposed model significantly outperforms the baseline and related work models.

Handwritten Signature Verification Method Using Convolutional Neural Network

Automatic signature verification methods play a significant role in providing a secure and authenticated handwritten signature in many applications, to prevent forgery problems, specifically institutions of finance, and transections of legal papers, etc. There are two types of handwritten signature verification methods: online verification (dynamic) and offline verification (static) methods. Besides, signature verification approaches can be categorized into two styles: writer dependent (WD), and writer independent (WI) styles. Offline signature verification methods demands a high representation features for the signature image. However, lots of studies have been proposed for WI offline signature verification. Yet, there is necessity to improve the overall accuracy measurements. Therefore, a proved solution in this paper is depended on deep learning via convolutional neural network (CNN) for signature verification and optimize the overall accuracy measurements. The introduced model is trained on English signature dataset. For model evaluation, the deployed model is utilized to make predictions on new data of Arabic signature dataset to classify whether the signature is real or forged. The overall obtained accuracy is 95.36\% based on validation dataset.

Arabic Music Genre Identification

Music Information Retrieval (MIR) is one data science application crucial for different tasks such as recommendation systems, genre identification, fingerprinting, and novelty assessment. Different Machine Learning techniques are utilised to analyse digital music records, such as clustering, classification, similarity scoring, and identifying various properties for the different tasks. Music is represented digitally using diverse transformations and is clustered and classified successfully for Western Music. However, Eastern Music poses a challenge, and some techniques have achieved success in clustering and classifying Turkish and Persian Music. This research presents an evaluation of machine learning algorithms' performance on pre-labelled Arabic Music with their Arabic genre (Maqam). The study introduced new data representations of the Arabic music dataset and identified the most suitable machine-learning methods and future enhancements.

Fine-grained hate speech detection in Arabic using transformer-based models

With the proliferation of social media platforms, characterized by features such as anonymity, user-friendly access, and the facilitation of online community building and discourse, the matter of detecting and monitoring hate speech has emerged as an increasingly formidable challenge for society, individuals, and researchers. Despite the crucial importance of hate speech detection task, the majority of work in this field has been conducted in English, with insufficient focus on other languages, particularly Arabic. Furthermore, most existing studies on Arabic hate speech detection have addressed this task as a binary classification problem, which is unreliable. Therefore, the aim of this study is to provide an enhanced model for detecting fine-grained hate speech in Arabic. To this end, three transformer-based models were evaluated to generate contextualized word embeddings from input sequence. Additionally, these models were combined with a bidirectional gated recurrent unit (BiGRU) layer to further improve the extracted semantic and context features. The experiments were conducted on an Arabic reference dataset provided by the open-source Arabic corpora and processing tools (OSACT-5) shared task. A comparative analysis indicates the efficiency of the proposed model over the baseline and related work models by achieving a macro F1-score of 61.68%.

An Ensemble Keyword Extraction Model for News Texts with Statistical and Graphical Features

Keyword extraction is an essential tool for many text mining applications such as automatic indexing, summarizing, classification, clustering and automatic filtering. Automated keyword extraction is essential as the daily text data to be reached and processed have increased tremendously over the Internet, e.g. millions of news articles are published daily online. In this paper, a novel ensemble model for automatic extraction of keywords from news articles is proposed. The proposed model handles keyword extraction as a sequence labeling task. Two sub-modules representing the statistical and graphical features by their calculated scores for each input token were combined in the token classification module. The Ensemble Token Classification module was trained and tested separately with the ensemble algorithms Random Forest, XgBoost, Decision Tree and Voting Classification. For training, we collected two news datasets from Kazakh and Russian news sites published in Cyrillic alphabet. We also collected an Arabic news dataset, ArabianNews. The performance of the model was also compared with the widely used 500N-KPCrowd dataset in the literature, which consists of English news content in Latin alphabet. The proposed model achieved the best performance with an [Formula: see text]-score of 0.71 and 0.86 on the 500N-KPCrowd and Russian datasets, respectively. We attained the best [Formula: see text]-score (0.97) with the KazakhNews and ArabianNews datasets.

AlFloos: a balanced dataset for the sixth issue of the Saudi Arabian currency banknote*

AbstractArtificial Intelligence, especially machine learning, is considered by many researchers the new electricity and an important building block of the fourth industrial revolution. Saudi Arabia is amongst the first nations worldwide to realize this importance and apply it in its school systems at various levels. This interest motivates researchers to create Arabic datasets for research and educational purposes, particularly with the popularity of English sources and the absence of Arabic sources. This study attempts to lessen this gap by creating for the first time ever a dataset for the sixth issue of the Saudi Arabian currency and analyzing it using the famous machine learning tools: Orange Data Mining, Google Teachable Machine, and Liner.ai, which require no coding. The importance of the study to the community is as follows: 1—It provides the first work ever to establish a balanced dataset for the sixth issue of the Saudi Arabian currency banknote, which contains images for the banknote denominations and the tabular data generated using deep learning. 2—It is the first scientific work that uses shallow machine learning and deep learning models to create good-performing models for classifying the sixth issue of the Saudi Arabian currency without coding. This work paves the way for researchers and those interested from various fields and backgrounds to develop machine learning applications to classify the sixth issue of the Saudi Arabian currency, especially in mobile phones or in microcontrollers, to inspire multiple IoT and Tiny machine learning applications like currency recognition and classification for special needs people, automatic counting and sorting of banknotes in banks and stores, and fake currency detection.

Beyond Event-Centric Narratives: Advancing Arabic Story Generation with Large Language Models and Beam Search

In the domain of automated story generation, the intricacies of the Arabic language pose distinct challenges. This study introduces a novel methodology that moves away from conventional event-driven narrative frameworks, emphasizing the restructuring of narrative constructs through sophisticated language models. Utilizing mBERT, our approach begins by extracting key story entities. Subsequently, XLM-RoBERTa and a BERT-based linguistic evaluation model are employed to direct beam search algorithms in the replacement of these entities. Further refinement is achieved through Low-Rank Adaptation (LoRA), which fine-tunes the extensive 3 billion-parameter BLOOMZ model specifically for generating Arabic narratives. Our methodology underwent thorough testing and validation, involving individual assessments of each submodel. The ROCStories dataset provided the training ground for our story entity extractor and new entity generator, and was also used in the fine-tuning of the BLOOMZ model. Additionally, the Arabic ComVE dataset was employed to train our commonsense evaluation model. Our extensive analyses yield crucial insights into the efficacy of our approach. The story entity extractor demonstrated robust performance with an F-score of 96.62%. Our commonsense evaluator reported an accuracy of 84.3%, surpassing the previous best by 3.1%. The innovative beam search strategy effectively produced entities that were linguistically and semantically superior to those generated using baseline models. Further subjective evaluations affirm our methodology’s capability to generate high-quality Arabic stories characterized by linguistic fluency and logical coherence.

More Than Syntaxes: Investigating Semantics to Zero-shot Cross-lingual Relation Extraction and Event Argument Role Labelling

Syntactic dependency structures are commonly utilized as language-agnostic features to solve the word order difference issues in zero-shot cross-lingual relation and event extraction tasks. However, while sentences in multiple forms can be employed to express the same meaning, the syntactic structure may vary considerably in specific scenarios. To fix this problem, we find semantics are rarely considered, which could provide a more consistent semantic analysis of sentences and be served as another bridge between different languages. Therefore, in this article, we introduce Syntax and Semantic Driven Network (SSDN) to equip syntax and semantic knowledge across languages simultaneously. Specifically, predicate–argument structures from semantic role labelling are explicitly incorporated into word representations. Then, a semantic-aware relational graph convolutional network and a transformer-based encoder are utilized to model both semantic dependency and syntactic dependency structures, respectively. Finally, a fusion module is introduced to integrate output representations adaptively. We conduct experiments on the widely used Automatic Content Extraction 2005 English, Chinese, and Arabic datasets. The evaluation results demonstrate that the proposed method achieves the state-of-the-art performance. Further study also indicates SSDN could produce robust representations that facilitate the transfer operations across languages.