Enabling empathetic behavior in Arabic dialogue agents is an important aspect of building human-like conversational models. While Arabic Natural Language Processing has seen significant advances in Natural Language Understanding (NLU) with language models such as AraBERT, Natural Language Generation (NLG) remains a challenge. The shortcomings of NLG encoder-decoder models are primarily due to the lack of Arabic datasets suitable to train NLG models such as conversational agents. To overcome this issue, we propose a transformer-based encoder-decoder initialized with AraBERT parameters. By initializing the weights of the encoder and decoder with AraBERT pre-trained weights, our model was able to leverage knowledge transfer and boost performance in response generation. To enable empathy in our conversational model, we train it using the ArabicEmpatheticDialogues dataset and achieve high performance in empathetic response generation. Specifically, our model achieved a low perplexity value of 17.0 and an increase in 5 BLEU points compared to the previous state-of-the-art model. Also, our proposed model was rated highly by 85 human evaluators, validating its high capability in exhibiting empathy while generating relevant and fluent responses in open-domain settings.

Advances in English language representation enabled a more sample-efficient pre-training task by Efficiently Learning an Encoder that Classifies Token Replacements Accurately (ELECTRA). Which, instead of training a model to recover masked tokens, it trains a discriminator model to distinguish true input tokens from corrupted tokens that were replaced by a generator network. On the other hand, current Arabic language representation approaches rely only on pretraining via masked language modeling. In this paper, we develop an Arabic language representation model, which we name AraELECTRA. Our model is pretrained using the replaced token detection objective on large Arabic text corpora. We evaluate our model on multiple Arabic NLP tasks, including reading comprehension, sentiment analysis, and named-entity recognition and we show that AraELECTRA outperforms current state-of-the-art Arabic language representation models, given the same pretraining data and with even a smaller model size.

Recently, pre-trained transformer-based architectures have proven to be very efficient at language modeling and understanding, given that they are trained on a large enough corpus. Applications in language generation for Arabic are still lagging in comparison to other NLP advances primarily due to the lack of advanced Arabic language generation models. In this paper, we develop the first advanced Arabic language generation model, AraGPT2, trained from scratch on a large Arabic corpus of internet text and news articles. Our largest model, AraGPT2-mega, has 1.46 billion parameters, which makes it the largest Arabic language model available. The mega model was evaluated and showed success on different tasks including synthetic news generation, and zero-shot question answering. For text generation, our best model achieves a perplexity of 29.8 on held-out Wikipedia articles. A study conducted with human evaluators showed the significant success of AraGPT2-mega in generating news articles that are difficult to distinguish from articles written by humans. We thus develop and release an automatic discriminator model with a 98% percent accuracy in detecting model-generated text. The models are also publicly available, hoping to encourage new research directions and applications for Arabic NLP.

The Arabic language is a morphologically rich language with relatively few resources and a less explored syntax compared to English. Given these limitations, Arabic Natural Language Processing (NLP) tasks like Sentiment Analysis (SA), Named Entity Recognition (NER), and Question Answering (QA), have proven to be very challenging to tackle. Recently, with the surge of transformers based models, language-specific BERT based models have proven to be very efficient at language understanding, provided they are pre-trained on a very large corpus. Such models were able to set new standards and achieve state-of-the-art results for most NLP tasks. In this paper, we pre-trained BERT specifically for the Arabic language in the pursuit of achieving the same success that BERT did for the English language. The performance of AraBERT is compared to multilingual BERT from Google and other state-of-the-art approaches. The results showed that the newly developed AraBERT achieved state-of-the-art performance on most tested Arabic NLP tasks. The pretrained araBERT models are publicly available on https://github.com/aub-mind/araBERT hoping to encourage research and applications for Arabic NLP.

The use of social media platforms has become more prevalent, which has provided tremendous opportunities for people to connect but has also opened the door for misuse with the spread of hate speech and offensive language. This phenomenon has been driving more and more people to more extreme reactions and online aggression, sometimes causing physical harm to individuals or groups of people. There is a need to control and prevent such misuse of online social media through automatic detection of profane language. The shared task on Offensive Language Detection at the OSACT4 has aimed at achieving state of art profane language detection methods for Arabic social media. Our team “BERTologists” tackled this problem by leveraging state of the art pretrained Arabic language model, AraBERT, that we augment with the addition of Multi-task learning to enable our model to learn efficiently from little data. Our Multitask AraBERT approach achieved the second place in both subtasks A & B, which shows that the model performs consistently across different tasks.

Conversational models have witnessed a significant research interest in the last few years with the advancements in sequence generation models. A challenging aspect in developing human-like conversational models is enabling the sense of empathy in bots, making them infer emotions from the person they are interacting with. By learning to develop empathy, chatbot models are able to provide human-like, empathetic responses, thus making the human-machine interaction close to human-human interaction. Recent advances in English use complex encoder-decoder language models that require large amounts of empathetic conversational data. However, research has not produced empathetic bots for Arabic. Furthermore, there is a lack of Arabic conversational data labeled with empathy. To address these challenges, we create an Arabic conversational dataset that comprises empathetic responses. However, the dataset is not large enough to develop very complex encoder-decoder models. To address the limitation of data scale, we propose a special encoder-decoder composed of a Long Short-Term Memory (LSTM) Sequence-to-Sequence (Seq2Seq) with Attention. The experiments showed success of our proposed empathy-driven Arabic chatbot in generating empathetic responses with a perplexity of 38.6, an empathy score of 3.7, and a fluency score of 3.92.

Arabizi is an informal written form of dialectal Arabic transcribed in Latin alphanumeric characters. It has a proven popularity on chat platforms and social media, yet it suffers from a severe lack of natural language processing (NLP) resources. As such, texts written in Arabizi are often disregarded in sentiment analysis tasks for Arabic. In this paper we describe the creation of a sentiment lexicon for Arabizi that was enriched with word embeddings. The result is a new Arabizi lexicon consisting of 11.3K positive and 13.3K negative words. We evaluated this lexicon by classifying the sentiment of Arabizi tweets achieving an F1-score of 0.72. We provide a detailed error analysis to present the challenges that impact the sentiment analysis of Arabizi.

Arabic is a complex language with limited resources which makes it challenging to produce accurate text classification tasks such as sentiment analysis. The utilization of transfer learning (TL) has recently shown promising results for advancing accuracy of text classification in English. TL models are pre-trained on large corpora, and then fine-tuned on task-specific datasets. In particular, universal language models (ULMs), such as recently developed BERT, have achieved state-of-the-art results in various NLP tasks in English. In this paper, we hypothesize that similar success can be achieved for Arabic. The work aims at supporting the hypothesis by developing the first Universal Language Model in Arabic (hULMonA - حلمنا meaning our dream), demonstrating its use for Arabic classifications tasks, and demonstrating how a pre-trained multi-lingual BERT can also be used for Arabic. We then conduct a benchmark study to evaluate both ULM successes with Arabic sentiment analysis. Experiment results show that the developed hULMonA and multi-lingual ULM are able to generalize well to multiple Arabic data sets and achieve new state of the art results in Arabic Sentiment Analysis for some of the tested sets.

This paper tackles the problem of open domain factual Arabic question answering (QA) using Wikipedia as our knowledge source. This constrains the answer of any question to be a span of text in Wikipedia. Open domain QA for Arabic entails three challenges: annotated QA datasets in Arabic, large scale efficient information retrieval and machine reading comprehension. To deal with the lack of Arabic QA datasets we present the Arabic Reading Comprehension Dataset (ARCD) composed of 1,395 questions posed by crowdworkers on Wikipedia articles, and a machine translation of the Stanford Question Answering Dataset (Arabic-SQuAD). Our system for open domain question answering in Arabic (SOQAL) is based on two components: (1) a document retriever using a hierarchical TF-IDF approach and (2) a neural reading comprehension model using the pre-trained bi-directional transformer BERT. Our experiments on ARCD indicate the effectiveness of our approach with our BERT-based reader achieving a 61.3 F1 score, and our open domain system SOQAL achieving a 27.6 F1 score.

While transfer learning for text has been very active in the English language, progress in Arabic has been slow, including the use of Domain Adaptation (DA). Domain Adaptation is used to generalize the performance of any classifier by trying to balance the classifier’s accuracy for a particular task among different text domains. In this paper, we propose and evaluate two variants of a domain adaptation technique: the first is a base model called Domain Adversarial Neural Network (DANN), while the second is a variation that incorporates representational learning. Similar to previous approaches, we propose the use of proxy A-distance as a metric to assess the success of generalization. We make use of ArSentDLEV, a multi-topic dataset collected from the Levantine countries, to test the performance of the models. We show the superiority of the proposed method in accuracy and robustness when dealing with the Arabic language.

While significant progress has been achieved for Opinion Mining in Arabic (OMA), very limited efforts have been put towards the task of Emotion mining in Arabic. In fact, businesses are interested in learning a fine-grained representation of how users are feeling towards their products or services. In this work, we describe the methods used by the team Emotion Mining in Arabic (EMA), as part of the SemEval-2018 Task 1 for Affect Mining for Arabic tweets. EMA participated in all 5 subtasks. For the five tasks, several preprocessing steps were evaluated and eventually the best system included diacritics removal, elongation adjustment, replacement of emojis by the corresponding Arabic word, character normalization and light stemming. Moreover, several features were evaluated along with different classification and regression techniques. For the 5 subtasks, word embeddings feature turned out to perform best along with Ensemble technique. EMA achieved the 1st place in subtask 5, and 3rd place in subtasks 1 and 3.

Nowadays, social media have become a platform where people can easily express their opinions and emotions about any topic such as politics, movies, music, electronic products and many others. On the other hand, politicians, companies, and businesses are interested in analyzing automatically people’s opinions and emotions. In the last decade, a lot of efforts has been put into extracting sentiment polarity from texts. Recently, the focus has expanded to also cover emotion recognition from texts. In this work, we expand an existing emotion lexicon, DepecheMood, by leveraging semantic knowledge from English WordNet (EWN). We create an expanded lexicon, EmoWordNet, consisting of 67K terms aligned with EWN, almost 1.8 times the size of DepecheMood. We also evaluate EmoWordNet in an emotion recognition task using SemEval 2007 news headlines dataset and we achieve an improvement compared to the use of DepecheMood. EmoWordNet is publicly available to speed up research in the field on http://oma-project.com.

Data generated on Twitter has become a rich source for various data mining tasks. Those data analysis tasks that are dependent on the tweet semantics, such as sentiment analysis, emotion mining, and rumor detection among others, suffer considerably if the tweet is not credible, not real, or spam. In this paper, we perform an extensive analysis on credibility of Arabic content on Twitter. We also build a classification model (CAT) to automatically predict the credibility of a given Arabic tweet. Of particular originality is the inclusion of features extracted directly or indirectly from the author’s profile and timeline. To train and test CAT, we annotated for credibility a data set of 9,000 Arabic tweets that are topic independent. CAT achieved consistent improvements in predicting the credibility of the tweets when compared to several baselines and when compared to the state-of-the-art approach with an improvement of 21% in weighted average F-measure. We also conducted experiments to highlight the importance of the user-based features as opposed to the content-based features. We conclude our work with a feature reduction experiment that highlights the best indicative features of credibility.

Opinion mining in Arabic is a challenging task given the rich morphology of the language. The task becomes more challenging when it is applied to Twitter data, which contains additional sources of noise, such as the use of unstandardized dialectal variations, the nonconformation to grammatical rules, the use of Arabizi and code-switching, and the use of non-text objects such as images and URLs to express opinion. In this paper, we perform an analytical study to observe how such linguistic phenomena vary across different Arab regions. This study of Arabic Twitter characterization aims at providing better understanding of Arabic Tweets, and fostering advanced research on the topic. Furthermore, we explore the performance of the two schools of machine learning on Arabic Twitter, namely the feature engineering approach and the deep learning approach. We consider models that have achieved state-of-the-art performance for opinion mining in English. Results highlight the advantages of using deep learning-based models, and confirm the importance of using morphological abstractions to address Arabic’s complex morphology.

While sentiment analysis in English has achieved significant progress, it remains a challenging task in Arabic given the rich morphology of the language. It becomes more challenging when applied to Twitter data that comes with additional sources of noise including dialects, misspellings, grammatical mistakes, code switching and the use of non-textual objects to express sentiments. This paper describes the “OMAM” systems that we developed as part of SemEval-2017 task 4. We evaluate English state-of-the-art methods on Arabic tweets for subtask A. As for the remaining subtasks, we introduce a topic-based approach that accounts for topic specificities by predicting topics or domains of upcoming tweets, and then using this information to predict their sentiment. Results indicate that applying the English state-of-the-art method to Arabic has achieved solid results without significant enhancements. Furthermore, the topic-based method ranked 1st in subtasks C and E, and 2nd in subtask D.

A significant portion of data generated on blogging and microblogging websites is non-credible as shown in many recent studies. To filter out such non-credible information, machine learning can be deployed to build automatic credibility classifiers. However, as in the case with most supervised machine learning approaches, a sufficiently large and accurate training data must be available. In this paper, we focus on building a public Arabic corpus of blogs and microblogs that can be used for credibility classification. We focus on Arabic due to the recent popularity of blogs and microblogs in the Arab World and due to the lack of any such public corpora in Arabic. We discuss our data acquisition approach and annotation process, provide rigid analysis on the annotated data and finally report some results on the effectiveness of our data for credibility classification.