Farasa: A Fast and Furious Segmenter for Arabic

Abdelali, Ahmed; Darwish, Kareem; Durrani, Nadir; Mubarak, Hamdy

doi:10.18653/v1/n16-3003

Cited by 248 publications

(186 citation statements)

References 13 publications

Supporting

Mentioning

186

Contrasting

Order By: Relevance

“…This work presents two open source state-of-theart POS tagging systems that are trained using standard ATB dataset (Maamouri et al, 2004) and evaluated on the WikiNews test set (Abdelali et al, 2016). In building the system we explored two approaches using Support Vector Machines (SVM) and Bidirectional Long Short-Term Memory (bi-LSTM).…”

Section: Resultsmentioning

confidence: 99%

“…Stem templates may conclusively have one POS tag (e.g., yCCC is always a V) or favor one tag over another (e.g., CCAC is more likely a NOUN than an ADJ). We used Farasa to determine the stem template (Abdelali et al, 2016). …”

Section: Tagging Cliticsmentioning

confidence: 99%

“…The clitics were obtained using either gold ATB segmentation or from the Farasa Segmenter (Abdelali et al, 2016).…”

Section: Bi-lstm Approachmentioning

confidence: 99%

“…However, combining both explicit features and embeddings together leads sub-optimal results. For testing, we employ the so-called "WikiNews" test set which is composed of freely available recent news articles in multiple genre (Abdelali et al, 2016). We are making all resultant systems available as opensource systems.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Arabic POS Tagging: Don't Abandon Feature Engineering Just Yet

Darwish¹,

Mubarak²,

Abdelali³

et al. 2017

Proceedings of the Third Arabic Natural Language Processing Workshop

Self Cite

View full text Add to dashboard Cite

This paper focuses on comparing between using Support Vector Machine based ranking (SVM Rank ) and Bidirectional LongShort-Term-Memory (bi-LSTM) neuralnetwork based sequence labeling in building a state-of-the-art Arabic part-ofspeech tagging system. Using SVM Rank leads to state-of-the-art results, but with a fair amount of feature engineering. Using bi-LSTM, particularly when combined with word embeddings, may lead to competitive POS-tagging results by automatically deducing latent linguistic features. However, we show that augmenting bi-LSTM sequence labeling with some of the features that we used for the SVM Rankbased tagger yields to further improvements. We also show that gains realized using embeddings may not be additive with the gains achieved due to features. We are open-sourcing both the SVM Rank and the bi-LSTM based systems for the research community.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Tagging Cliticsmentioning

confidence: 99%

“…The clitics were obtained using either gold ATB segmentation or from the Farasa Segmenter (Abdelali et al, 2016).…”

Section: Bi-lstm Approachmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Arabic POS Tagging: Don't Abandon Feature Engineering Just Yet

Darwish¹,

Mubarak²,

Abdelali³

et al. 2017

Proceedings of the Third Arabic Natural Language Processing Workshop

Self Cite

View full text Add to dashboard Cite

show abstract

“…• The prefix wa "and" is separated from words by using Farasa toolkit (Abdelali et al, 2016) and all other prefixes: b, f, Al, k, l and s are concatenated to words.…”

Section: Data Normalizationmentioning

confidence: 99%

An enhanced automatic speech recognition system for Arabic

Menacer¹,

Mella²,

Fohr³

et al. 2017

Proceedings of the Third Arabic Natural Language Processing Workshop

View full text Add to dashboard Cite

Automatic speech recognition for Arabic is a very challenging task. Despite all the classical techniques for Automatic Speech Recognition (ASR), which can be efficiently applied to Arabic speech recognition, it is essential to take into consideration the language specificities to improve the system performance. In this article, we focus on Modern Standard Arabic (MSA) speech recognition. We introduce the challenges related to Arabic language, namely the complex morphology nature of the language and the absence of the short vowels in written text, which leads to several potential vowelization for each graphemes, which is often conflicting. We develop an ASR system for MSA by using Kaldi toolkit. Several acoustic and language models are trained. We obtain a Word Error Rate (WER) of 14.42 for the baseline system and 12.2 relative improvement by rescoring the lattice and by rewriting the output with the right hamoza above or below Alif.

show abstract