In this paper, we present AfricaPOS, the largest part-of-speech (POS) dataset for 20 typologically diverse African languages. We discuss the challenges in annotating POS for these languages using the universal dependencies (UD) guidelines. We conducted extensive POS baseline experiments using both conditional random field and several multilingual pre-trained language models. We applied various cross-lingual transfer models trained with data available in the UD. Evaluating on the AfricaPOS dataset, we show that choosing the best transfer language(s) in both single-source and multi-source setups greatly improves the POS tagging performance of the target languages, in particular when combined with parameter-fine-tuning methods. Crucially, transferring knowledge from a language that matches the language family and morphosyntactic properties seems to be more effective for POS tagging in unseen languages.

African languages are spoken by over a billion people, but they are under-represented in NLP research and development. Multiple challenges exist, including the limited availability of annotated training and evaluation datasets as well as the lack of understanding of which settings, languages, and recently proposed methods like cross-lingual transfer will be effective. In this paper, we aim to move towards solutions for these challenges, focusing on the task of named entity recognition (NER). We present the creation of the largest to-date human-annotated NER dataset for 20 African languages. We study the behaviour of state-of-the-art cross-lingual transfer methods in an Africa-centric setting, empirically demonstrating that the choice of source transfer language significantly affects performance. While much previous work defaults to using English as the source language, our results show that choosing the best transfer language improves zero-shot F1 scores by an average of 14% over 20 languages as compared to using English.

In this paper, we propose two neural machine translation (NMT) systems (French-to-Wolof and Wolof-to-French) based on sequence-to-sequence with attention and Transformer architectures. We trained our models on the parallel French-Wolof corpus (Nguer et al., 2020) of about 83k sentence pairs. Because of the low-resource setting, we experimented with advanced methods for handling data sparsity, including subword segmentation, backtranslation and the copied corpus method. We evaluate the models using BLEU score and find that the transformer outperforms the classic sequence-to-sequence model in all settings, in addition to being less sensitive to noise. In general, the best scores are achieved when training the models on subword-level based units. For such models, using backtranslation proves to be slightly beneficial in low-resource Wolof to high-resource French language translation for the transformer-based models. A slight improvement can also be observed when injecting copied monolingual text in the target language. Moreover, combining the copied method data with backtranslation leads to a slight improvement of the translation quality.

This paper describes a methodology for syntactic knowledge transfer between high-resource languages to extremely low-resource languages. The methodology consists in leveraging multilingual BERT self-attention model pretrained on large datasets to develop a multilingual multi-task model that can predict Universal Dependencies annotations for three African low-resource languages. The UD annotations include universal part-of-speech, morphological features, lemmas, and dependency trees. In our experiments, we used multilingual word embeddings and a total of 11 Universal Dependencies treebanks drawn from three high-resource languages (English, French, Norwegian) and three low-resource languages (Bambara, Wolof and Yoruba). We developed various models to test specific language combinations involving contemporary contact languages or genetically related languages. The results of the experiments show that multilingual models that involve high-resource languages and low-resource languages with contemporary contact between each other can provide better results than combinations that only include unrelated languages. As far genetic relationships are concerned, we could not draw any conclusion regarding the impact of language combinations involving the selected low-resource languages, namely Wolof and Yoruba.

In this paper, we report efforts towards the acquisition and construction of a bilingual parallel corpus between French and Wolof, a Niger-Congo language belonging to the Northern branch of the Atlantic group. The corpus is constructed as part of the SYSNET3LOc project. It currently contains about 70,000 French-Wolof parallel sentences drawn on various sources from different domains. The paper discusses the data collection procedure, conversion, and alignment of the corpus as well as it’s application as training data for neural machine translation. In fact, using this corpus, we were able to create word embedding models for Wolof with relatively good results. Currently, the corpus is being used to develop a neural machine translation model to translate French sentences into Wolof.

This paper reports on a parsing system for Wolof based on the LFG formalism. The parser covers core constructions of Wolof, including noun classes, cleft, copula, causative and applicative sentences. It also deals with several types of coordination, including same constituent coordination, asymmetric and asyndetic coordination. The system uses a cascade of finite-state transducers for word tokenization and morphological analysis as well as various lexicons. In addition, robust parsing techniques, including fragmenting and skimming, are used to optimize grammar coverage. Parsing coverage is evaluated by running test-suites of naturally occurring Wolof sentences through the parser. The evaluation of parsing coverage reveals that 72.72% of the test sentences receive full parses; 27.27% receive partial parses. To measure accuracy, the parsed sentences are disambiguated manually using an incremental parsebanking approach based on discriminants. The evaluation of parsing quality reveals that the parser achieves 67.2% recall, 92.8% precision and an f-score of 77.9%.

This paper reports on a systematic approach for deriving Universal Dependencies from LFG structures. The conversion starts with a step-wise transformation of the c-structure, combining part-of-speech (POS) information and the embedding path to determine the true head of dependency structures. The paper discusses several issues faced by existing algorithms when applied on Wolof and presents the strategies used to account for these issues. An experimental evaluation indicated that our approach was able to generate the correct output in more than 90% of the cases, leading to a substantial improvement in conversion accuracy compared to the previous models.

This paper presents a method for greatly reducing parse times in LFG by integrating a Constraint Grammar parser into a probabilistic context-free grammar. The CG parser is used in the pre-processing phase to reduce morphological and lexical ambiguity. Similarly, the c-structure pruning mechanism of XLE is used in the parsing phase to discard low-probability c-structures, before f-annotations are solved. The experiment results show a considerable increase in parsing efficiency and robustness in the annotation of Wolof running text. The Wolof CG parser indicated an f-score of 90% for morphological disambiguation and a speedup of ca. 40%, while the c-structure pruning method increased the speed of the Wolof grammar by over 36%. On a small amount of data, CG disambiguation and c-structure pruning allowed for a speedup of 58%, however with a substantial drop in parse accuracy of 3.62.

This paper reports on the design and implementation of a morphological analyzer for Wolof. The main motivation for this work is to obtain a linguistically motivated tool using finite-state techniques. The finite-state technology is especially attractive in dealing with human language morphologies. Finite-state transducers (FST) are fast, efficient and can be fully reversible, enabling users to perform analysis as well as generation. Hence, I use this approach to construct a new FST tool for Wolof, as a first step towards a computational grammar for the language in the Lexical Functional Grammar framework. This article focuses on the methods used to model complex morphological issues and on developing strategies to limit ambiguities. It discusses experimental evaluations conducted to assess the performance of the analyzer with respect to various statistical criteria. In particular, I also wanted to create morphosyntactically annotated resources for Wolof, obtained by automatically analyzing text corpora with a computational morphology.

In this paper, we report on the design of a part-of-speech-tagset for Wolof and on the creation of a semi-automatically annotated gold standard. In order to achieve high-quality annotation relatively fast, we first generated an accurate lexicon that draws on existing word and name lists and takes into account inflectional and derivational morphology. The main motivation for the tagged corpus is to obtain data for training automatic taggers with machine learning approaches. Hence, we took machine learning considerations into account during tagset design and we present training experiments as part of this paper. The best automatic tagger achieves an accuracy of 95.2% in cross-validation experiments. We also wanted to create a basis for experimenting with annotation projection techniques, which exploit parallel corpora. For this reason, it was useful to use a part of the Bible as the gold standard corpus, for which sentence-aligned parallel versions in many languages are easy to obtain. We also report on preliminary experiments exploiting a statistical word alignment of the parallel text.