Spoken Tunisian Arabic Corpus “STAC”: Transcription and Annotation

Nowadays, Tunisian Dialect (TD) is widely used in social media such as Facebook and Twitter to communicate and express opinions. However, the absence of a TD translator prevents comprehension and communication for non-Tunisian users. Thus, Tunisian translator becomes an important task to overcome these problems. In this article, our main objective is to create a translator from TD to Modern Standard Arabic (MSA). In this context, we are carrying out an in-depth linguistic study in order to better choose the translation rules which are applicable on the TD sentences. This study guarantees the best quality to obtain MSA sentences. Indeed, our proposed method is based on a linguistic approach. In addition, the method consists of the elaboration of a set of dictionaries and the construction of inflectional, morphological, and syntactic grammars using finite-state transducers. Then, our method is implemented and tested with new technologies provided by the NooJ linguistic platform. To evaluate our constructed tool, we apply it to two different test corpora containing more than 15,000 words. The obtained results are promising and do highlight our proposed method.

show abstract

“…It holds 1600 sentences containing 8627 words. The second corpus is the Zribi's corpus 35 . It is collected from TV channels and radio stations.…”

Section: Discussionmentioning

confidence: 99%

“…It recognizes different phrases and compounds. The unrecognized words in Zribi's corpus 35 are almost related to the use of a lot of MSA words. Our Tunisian morphological analyzer therefore cannot treat these words as the word “ >afoDalo ” (the best).…”

Section: Discussionmentioning

confidence: 99%

Translation system from Tunisian Dialect to Modern Standard Arabic

Torjmen

Haddar

2021

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Levantine Arabic has also received a lot of attention, as in the creation of the Levantine Arabic Treebank (LATB) ( Maamouri et al, 2006 ), including 27,000 words in Jordanian Arabic. Some efforts were made for Tunisian ( Masmoudi et al, 2014 ; Zribi et al, 2015 ), and Algerian ( Smaıli et al, 2014 ). For Gulf Arabic, the Gumar corpus ( Khalifa et al, 2016a ) consists of 1,200 documents written in Gulf Arabic dialects from different forum novels available online ( https://nyuad.nyu.edu/en/research/centers-labs-and-projects/computational-approaches-to-modeling-language-lab/resources.html ).…”

Section: Related Researchmentioning

confidence: 99%

AraCust: a Saudi Telecom Tweets corpus for sentiment analysis

Almuqren

Cristea

2021

PeerJ Computer Science

View full text Add to dashboard Cite

Comparing Arabic to other languages, Arabic lacks large corpora for Natural Language Processing (Assiri, Emam & Al-Dossari, 2018; Gamal et al., 2019). A number of scholars depended on translation from one language to another to construct their corpus (Rushdi-Saleh et al., 2011). This paper presents how we have constructed, cleaned, pre-processed, and annotated our 20,0000 Gold Standard Corpus (GSC) AraCust, the first Telecom GSC for Arabic Sentiment Analysis (ASA) for Dialectal Arabic (DA). AraCust contains Saudi dialect tweets, processed from a self-collected Arabic tweets dataset and has been annotated for sentiment analysis, i.e.,manually labelled (k=0.60). In addition, we have illustrated AraCust’s power, by performing an exploratory data analysis, to analyse the features that were sourced from the nature of our corpus, to assist with choosing the right ASA methods for it. To evaluate our Golden Standard corpus AraCust, we have first applied a simple experiment, using a supervised classifier, to offer benchmark outcomes for forthcoming works. In addition, we have applied the same supervised classifier on a publicly available Arabic dataset created from Twitter, ASTD (Nabil, Aly & Atiya, 2015). The result shows that our dataset AraCust outperforms the ASTD result with 91% accuracy and 89% F1avg score. The AraCust corpus will be released, together with code useful for its exploration, via GitHub as a part of this submission.

show abstract

“…Several works have been carried out to create parallel and annotated corpora, notably by Graja et al (2013) and Zribi et al (2015). The work of Graja et al (2013) focused on the semantic annotation of spoken TD, using a discriminative model based on conditional random fields (CRFs).…”

Section: Related Workmentioning

confidence: 99%

“…The work of Graja et al (2013) focused on the semantic annotation of spoken TD, using a discriminative model based on conditional random fields (CRFs). As for Zribi et al (2015) they developed two corpora, namely a written corpus segmented in sentences, whose words are segmented and annotated by lemmas, gender, number, person, voice, labels, etc., as well as an oral corpus annotated by the different types of disfluencies. Mdhaffar et al (2017) collected a corpus, called Tunisian sentiment analysis corpus (TSAC) from Facebook, which they manually annotated with positive and negative polarities.…”

Section: Related Workmentioning

confidence: 99%

Contributions to the automatic processing of the user-generated Tunisian dialect on the social web

et al. 2020

View full text Add to dashboard Cite

With the growing use of social media in the Arab world, Arabic dialects are rapidly spreading on the web, leading to a growing interest from NLP researchers. These dialects are however, still under-resourced languages which is a major obstacle to their study and processing. In this paper, we focus on the automatic processing of the user-generated Tunisian dialect (TD) on the social web and propose an approach that aids to automatically generate TD language resources. This approach exploits the large amounts of textual productions on the social web to extract and generate dialectal content. It is based on two main NLP components, namely the TD identification and the TD transliteration. A machine learning approach using conditional random fields is proposed for implementing these two components and reached an accuracy of 87.45 for the TD identification and 90.49 for the automatic generation of dialectal contents by transliteration.

show abstract

Spoken Tunisian Arabic Corpus “STAC”: Transcription and Annotation

Cited by 27 publications

References 11 publications

Translation system from Tunisian Dialect to Modern Standard Arabic

Translation system from Tunisian Dialect to Modern Standard Arabic

AraCust: a Saudi Telecom Tweets corpus for sentiment analysis

Contributions to the automatic processing of the user-generated Tunisian dialect on the social web

Contact Info

Product

Resources

About