Investigating an approach for low resource language dataset creation, curation and classification: Setswana and Sepedi

Marivate, Vukosi; Sefara, Tshephisho Joseph; Chabalala, Vongani; Makhaya, Keamogetswe; Mokgonyane, Tumisho Billson; Mokoena, Rethabile; Modupe, Abiodun

doi:10.48550/arxiv.2003.04986

Cited by 6 publications

(19 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A trivial attempt to address this imbalance is to simply redirect research attention to the so-called resourced-deprived languages. However, such initiatives were less successful due to the high cost of creating, curating, and annotating quality datasets for low-resourced languages 11 . Instead, a research focus on using the morpho-syntactic embedding spaces of these available highresourced languages to supplement language-generic knowledge encoded in embedding spaces of low-resourced languages became an active research alternative: a field of extracting cross-language (CL) embeddings.…”

Section: Introductionmentioning

confidence: 99%

“…This means that the majority of the languages unable to meet these requirements are still not alleviated from the challenge of under-representation 45 . South African languages reserving a seat amongst the cohort unable to meet the aforementioned prerequisites 10,11,46 . For this, we aim to explore News Headlines Classification (NHC) 11 and Named Entity Recognition (NER) 47 downstream tasks on four agglutinative of the 11 official South African languages: Isixhosa (languages of the Nguni tribe), Sesotho, Setswana, and Sepedi (three languages of the Sotho-Tswana language family).…”

Section: Introductionmentioning

confidence: 99%

“…South African languages reserving a seat amongst the cohort unable to meet the aforementioned prerequisites 10,11,46 . For this, we aim to explore News Headlines Classification (NHC) 11 and Named Entity Recognition (NER) 47 downstream tasks on four agglutinative of the 11 official South African languages: Isixhosa (languages of the Nguni tribe), Sesotho, Setswana, and Sepedi (three languages of the Sotho-Tswana language family). The unique contributions of this study are organized into the following three paradigms:…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Zero-Shot Transfer Learning using Affix and Correlated Cross-Lingual Embeddings.

Modupe

Sindane

Marivate

2023

Preprint

View full text Add to dashboard Cite

Learning morphologically supplemented embedding spaces using cross-lingual models has become an active area of research and facilitated many research breakthroughs in various applications such as machine translation, named entity recognition, document classification, and natural language inference. However, the field has not become customary for Southern African low-resourced languages. In this paper, we present, evaluate and benchmark a cohort of cross-lingual embeddings for the English-Southern African languages on two classification tasks: News Headlines Classification (NHC) and Named Entity Recognition (NER). Our methodology considers four agglutinative languages from the eleven official South African languages: Isixhosa, Sepedi, Sesotho, and Setswana. Canonical correlation analyses and VecMap are the two cross-lingual alignment strategies adopted for this study. Monolingual embeddings used in this work are Glove (source), and FastText (source and target) embeddings. Our results indicate that with enough comparable corpora, we can develop strong inter-joined representations between English and the considered Southern African languages. More specifically, the best zero-shot transfer results on the available Setswana NHC dataset were achieved using canonically correlated embeddings with Multi-layered perceptron as the training model (54.5% accuracy). Furthermore, our NER best performance was achieved using canonically correlated cross-lingual embeddings with Conditional Random Fields as the training model (96.4% F1 score). Collectively, this study’s results were competitive with the benchmarks of the explored NHC and NER datasets, on both zero-short NHC and NER tasks with our advantage being the use of very minimal resources.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Zero-Shot Transfer Learning using Affix and Correlated Cross-Lingual Embeddings.

Modupe

Sindane

Marivate

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The development, training, and evaluation A word embedding trained on South African news data of word embedding models must therefore be context-specific. Examples of word embeddings linked to a certain domain are: the NukeBERT model ( Jain et al, 2020) that is trained on texts from the nuclear and atomic energy section; specialised embeddings for finance (Theil et al, 2020); and embeddings trained on certain languages, such as Setswana and Sepedi (Marivate et al, 2020) or Croatian (Svoboda & Beliga, 2017).…”

Section: Introductionmentioning

confidence: 99%

“…The word embedding we generated is publicly available via a github repository. 1 It is, to the best of our knowledge, the first publicly available word embedding trained on South Africa news article data, and thus forms a valuable addition to the field of NLP in African contexts (Marivate et al, 2020). The embedding will allow researchers to investigate the meanings of numerous words from within a South African context and to seek answers to culturally or politically oriented South African research questions-such as, to give but one small example, how the African National Congress (ANC) and Democratic Alliance (DA) relate to terms such as "corruption" and "white monopoly capital".…”

Section: Introductionmentioning

confidence: 99%

A word embedding trained on South African news data

Mafunda

Schuld

Durrheim

et al. 2022

Afr. j. inf. commun. (Online)

View full text Add to dashboard Cite

This article presents results from a study that developed and tested a word embedding trained on a dataset of South African news articles. A word embedding is an algorithm-generated word representation that can be used to analyse the corpus of words that the embedding is trained on. The embedding on which this article is based was generated using the Word2Vec algorithm, which was trained on a dataset of 1.3 million African news articles published between January 2018 and March 2021, containing a vocabulary of approximately 124,000 unique words. The efficacy of this Word2Vec South African news embedding was then tested, and compared to the efficacy provided by the globally used GloVe algorithm. The testing of the local Word2Vec embedding showed that it performed well, with similar efficacy to that provided by GloVe. The South African news word embedding generated by this study is freely available for public use.

show abstract

Building Text and Speech Benchmark Datasets and Models for Low‐Resourced East African Languages: Experiences and Lessons

Nakatumba‐Nabende,

Babirye,

Nabende

et al. 2024

Applied AI Letters

View full text Add to dashboard Cite

Africa has over 2000 languages; however, those languages are not well represented in the existing natural language processing ecosystem. African languages lack essential digital resources to effectively engage in advancing language technologies. There is a need to generate high‐quality natural language processing resources for low‐resourced African languages. Obtaining high‐quality speech and text data is expensive and tedious because it can involve manual sourcing and verification of data sources. This paper discusses the process taken to curate and annotate text and speech datasets for five East African languages: Luganda, Runyankore‐Rukiga, Acholi, Lumasaba, and Swahili. We also present results obtained from baseline models for machine translation, topic modeling and classification, sentiment classification, and automatic speech recognition tasks. Finally, we discuss the experiences, challenges, and lessons learned in creating the text and speech datasets.

show abstract

Investigating an approach for low resource language dataset creation, curation and classification: Setswana and Sepedi

Cited by 6 publications

References 0 publications

Zero-Shot Transfer Learning using Affix and Correlated Cross-Lingual Embeddings.

Zero-Shot Transfer Learning using Affix and Correlated Cross-Lingual Embeddings.

A word embedding trained on South African news data

Building Text and Speech Benchmark Datasets and Models for Low‐Resourced East African Languages: Experiences and Lessons

Contact Info

Product

Resources

About