Hasil untuk "African languages and literature"

Nikolaos Papaspanos

Background: Public health recommendations and governmental measures during the COVID-19 pandemic have resulted in numerous restrictions on daily living including social distancing, isolation and home confinement. While these measures are imperative to abate the spreading of COVID-19, the impact of these restrictions on health behaviours and lifestyles at home is undefined. Therefore, an international online survey was launched in April 2020, in seven languages, to elucidate the behavioural and lifestyle consequences of COVID-19 restrictions. This report presents the results from the first thousand responders on physical activity (PA) and nutrition behaviours. Subjects and methods: Following a structured review of the literature, the «Effects of home Confinement on multiple Lifestyle Behaviours during the COVID-19 outbreak (ECLB-COVID19)» Electronic survey was designed by a steering group of multidisciplinary scientists and academics. The survey was uploaded and shared on the Google online survey platform. Thirty-five research organisations from Europe, North Africa, Western Asia and the Americas promoted the survey in English, German, French, Arabic, Spanish, Portuguese and Slovenian languages. Questions were presented in a differential format, with questions related to responses «before» and «during» confinement conditions. Results: 1047 replies (54% women) from Asia (36%), Africa (40%), Europe (21%) and other (3%) were included in the analysis. The COVID-19 home confinement had a negative effect on all PA intensity levels (vigorous, moderate, walking and overall). Additionally, daily sitting time increased from 5 to 8 h per day. Food consumption and meal patterns (the type of food, eating out of control, snacks between meals, number of main meals) were more unhealthy during confinement, with only alcohol binge drinking decreasing significantly. Summary: While isolation is a necessary measure to protect public health, results indicate that it alters physical activity and eating behaviours in a health compromising direction. A more detailed analysis of survey data will allow for a segregation of these responses in different age groups, countries and other subgroups, which will help develop interventions to mitigate the negative lifestyle behaviours that have manifested during the COVID-19 confinement.

M. Kramer, R. Kakuma

Rockie Sibanda, Lina P. Tshehla

Background: Given the lack of research into English language instruction in multilingual contexts, this study explored the switch from mother tongue to English in a South African township school.Aim: This study aims to find out how teachers and parents view the implementation of English as a medium of instruction.Setting: The study was conducted at a multilingual township primary school that implemented an English-medium instruction policy from the foundation phase.Methods: A case study approach was applied to this qualitative research study. Semi-structured interviews were conducted with four foundation phase teachers and three parents of learners. In addition, observations of the classes of the selected teachers were made, and the data were analysed thematically.Results: The findings suggest that the English medium of instruction poses barriers to effective learning and teaching at the foundation phase. For most township learners, English is not the dominant language in their everyday interactions, and they have limited contact with the language outside the classroom. In addition, most teachers struggle to use English as a medium of instruction.Conclusion: Although some schools have shifted to an English medium policy, the decision appears ill-conceived because its success is insignificant in South Africa, where English is the first additional language for most learners and teachers.Contribution: The findings offer a research framework formulated by integrating past literature, and a theoretical framework for understanding the English medium of instruction where learners are exposed to multiple languages.

Muhammad Khamis Dauda, U. Musa, Hassanin M. Al-Barhamtoshy

The Hausa language, spoken by over 64 million people, is a vital part of West Africa's cultural and social fabric. Despite its importance, handwritten Hausa script recognition has little concern due to its tonal nature, unique characters, and limited datasets. This study tackles these challenges by creating a dataset and evaluating two models commonly used in handwritten recognition: a Convolutional Neural Network (CNN) and Tesseract Optical Character Recognition (OCR). The custom CNN, trained on data from 30 volunteers with varied writing styles, achieved an accuracy of 96%, with strong precision, recall, and F1-score metrics. In contrast, Tesseract OCR struggled, particularly with Hausa-specific characters like Ɓ, Ƙ, Ƴ, and ɗ, achieving lowest accuracy for these classes. Although it had non-zero recognition in some characters, its performance was inconsistent, indicating difficulties in generalization. This work provides a scalable solution for handwritten Hausa character recognition, contributing to research on low-resource African languages and paving the way for better access to Hausa handwritten literature and integration into natural language processing applications.

Boqi Chen, Ou Wei, Bingzhou Zheng et al.

Graph model generation from natural language description is an important task with many applications in software engineering. With the rise of large language models (LLMs), there is a growing interest in using LLMs for graph model generation. Nevertheless, LLM-based graph model generation typically produces partially correct models that suffer from three main issues: (1) syntax violations: the generated model may not adhere to the syntax defined by its metamodel, (2) constraint inconsistencies: the structure of the model might not conform to some domain-specific constraints, and (3) inaccuracy: due to the inherent uncertainty in LLMs, the models can include inaccurate, hallucinated elements. While the first issue is often addressed through techniques such as constraint decoding or filtering, the latter two remain largely unaddressed. Motivated by recent self-consistency approaches in LLMs, we propose a novel abstraction-concretization framework that enhances the consistency and quality of generated graph models by considering multiple outputs from an LLM. Our approach first constructs a probabilistic partial model that aggregates all candidate outputs and then refines this partial model into the most appropriate concrete model that satisfies all constraints. We evaluate our framework on several popular open-source and closed-source LLMs using diverse datasets for model generation tasks. The results demonstrate that our approach significantly improves both the consistency and quality of the generated graph models.

William J. Bowman

We present the design and implementation of a macro-embedding of a family of compiler intermediate languages, from a Scheme-like language to x86-64, into Racket. This embedding is used as part of a testing framework for a compilers course to derive interpreters for all the intermediate languages. The embedding implements features including safe, functional abstractions as well as unsafe assembly features, and the interactions between the two at various intermediate stages. This paper aims to demonstrate language-oriented techniques and abstractions for implementing (1) a large family of languages and (2) interoperability between low- and high-level languages. The primary strength of this approach is the high degree of code reuse and interoperability compared to implementing each interpreter separately. The design emphasizes modularity and compositionality of an open set of language features by local macro expansion into a single host language, rather than implementing a language pre-defined by a closed set of features. This enables reuse from both the host language (Racket) and between intermediate languages, and enables interoperability between high- and low-level features, simplifying development of the intermediate language semantics. It also facilitates extending or redefining individual language features in intermediate languages, and exposing multiple interfaces to the embedded languages.

Ruben Becker, Giuseppa Castiglione, Giovanna D'Agostino et al.

The notion of Wheeler languages is rooted in the Burrows-Wheeler transform (BWT), one of the most central concepts in data compression and indexing. The BWT has been generalized to finite automata, the so-called Wheeler automata, by Gagie et al. [Theor. Comput. Sci. 2017]. Wheeler languages have subsequently been defined as the class of regular languages for which there exists a Wheeler automaton accepting them. Besides their advantages in data indexing, these Wheelerlanguages also satisfy many interesting properties from a language theoretic point of view [Alanko et al., Inf. Comput. 2021]. A characteristic yet unsatisfying feature of Wheeler languages however is that their definition depends on a fixed order of the alphabet. In this paper we introduce the Universally Wheeler languages UW, i.e., the regular languages that are Wheeler with respect to all orders of a given alphabet. Our first main contribution is to relate UW to some very well known regular language classes. We first show that the Striclty Locally Testable languages are strictly included in UW. After noticing that UW is not closed under taking the complement, we prove that the class of languages for which both the language and its complement are in UW exactly coincides with those languages that are Definite or Reverse Definite. Secondly, we prove that deciding if a regular language given by a DFA is in UW can be done in quadratic time. We also show that this is optimal unless the Strong Exponential Time Hypothesis (SETH) fails.

Sam Stites, John M. Li, Steven Holtzen

There are many different probabilistic programming languages that are specialized to specific kinds of probabilistic programs. From a usability and scalability perspective, this is undesirable: today, probabilistic programmers are forced up-front to decide which language they want to use and cannot mix-and-match different languages for handling heterogeneous programs. To rectify this, we seek a foundation for sound interoperability for probabilistic programming languages: just as today's Python programmers can resort to low-level C programming for performance, we argue that probabilistic programmers should be able to freely mix different languages for meeting the demands of heterogeneous probabilistic programming environments. As a first step towards this goal, we introduce \textsc{MultiPPL}, a probabilistic multi-language that enables programmers to interoperate between two different probabilistic programming languages: one that leverages a high-performance exact discrete inference strategy, and one that uses approximate importance sampling. We give a syntax and semantics for \textsc{MultiPPL}, prove soundness of its inference algorithm, and provide empirical evidence that it enables programmers to perform inference on complex heterogeneous probabilistic programs and flexibly exploits the strengths and weaknesses of two languages simultaneously.%

H. Kebede, H. Gesesew, A. Gebremedhin et al.

Despite the fact that Sub-Saharan Africa bears a disproportionate burden of armed conflicts and HIV infection, there has been inadequate synthesis of the impact of armed conflict on HIV treatment outcomes. We summarized the available evidence on the impact of armed conflicts on HIV treatment outcomes in Sub-Saharan Africa from 2002 to 2022. We searched four databases; MEDLINE, PubMed, CINHAL, and Scopus. We also explored grey literature sources and reviewed the bibliographies of all articles to identify any additional relevant studies. We included quantitative studies published in English from January 1, 2002 to December 30, 2022 that reported on HIV treatment outcomes for patients receiving antiretroviral therapy (ART) in conflict and post-conflict areas, IDP centers, or refugee camps, and reported on their treatment outcomes from sub-Saharan Africa. Studies published in languages other than English, reporting on non-ART patients and reporting on current or former military populations were excluded. We used EndNote X9 and Covidence to remove duplicates, extracted data using JBI-MAStARI, assessed risk of bias using AHRQ criteria, reported results using PRISMA checklist, and determined Statistical heterogeneity using Cochran Q test and Higgins I2, R- and RevMan-5 software were used for meta-analysis. The review included 16 studies with participant numbers ranging from 102 to 2572. Lost To Follow-Up (LTFU) percentages varied between 5.4% and 43.5%, virologic non-suppression rates ranged from 25 to 33%, adherence rates were over 88%, and mortality rates were between 4.2% and 13%. A pooled meta-analysis of virologic non-suppression rates from active conflict settings revealed a non-suppression rate of 30% (0.30 (0.26–0.33), I2 = 0.00%, p = 0.000). In contrast, a pooled meta-analysis of predictors of loss to follow-up (LTFU) from post-conflict settings identified a higher odds ratio for females compared to males (1.51 (1.05, 2.17), I2 = 0%, p = 0.03). The review highlights a lack of research on the relationship between armed conflicts and HIV care outcomes in SSA. The available documents lack quality of designs and data sources, and the depth and diversity of subjects covered.

Victor H. Mlambo

This paper examines the challenges of protecting refugees in sub-Saharan Africa. Reviewing available and relevant literature on refugees and their displacement in this region, it argues that people are compelled to flee their homes and seek refuge across international borders due to poverty, instability, conflict, and climate-related emergencies. They are then placed in overcrowded camps, frequently for extended periods. In theory, African refugees can access one of the most progressive protection schemes in the world. In practice, they confront insurmountable obstacles to their human rights, including coerced return, prejudice, arbitrary detention and arrest, limitations on their freedoms of speech and movement, and violations of their social and economic rights. Refugee law, which was created in the language of human rights and applies to people who are already highly vulnerable, disappears from view amid the poor human rights records of many African countries. From an in-depth argument and reflection on the literature, considering the differing voices and arguments, this paper communicates how, given the rise in extremism and political instability in sub-Saharan Africa, regional security is at risk if the inadequate protection given to refugee camps is not addressed.

Fabian A.W. Meyers, Cornelia Smith, Madoda Cekiso

Teachers in the current digital era are required to integrate Information and Communication Technology (ICT) in their daily teaching and must replace their traditional methods with modern tools and facilities. This is because ICT provides a dynamic and proactive teaching and learning environment. Consequently, the current study sought to establish how English First Additional Language (EFAL) Technical and Vocational Education and Training (TVET) college first-year students understood face-to-face instruction and blended learning (BL) environments for academic writing. The study was qualitative in nature and a case study design was followed. Twelve purposively selected first-year students were involved in semi-structured interviews as part of data collection. In this study constructivism was used as theoretical framework with reference to BL and academic writing. The findings of the study revealed that most students were in favour of the face-to-face learning mode because of its advantages in their learning context. Those who were not in favour of BL posited that it had the potential to facilitate inequality among students as it was likely to benefit only those who could afford to buy data. The findings further revealed that participants believed that the combination of both face-to-face and online learning modes may be conducive to the context of learning academic writing. They contend that the two types of learning are inextricably linked. Contribution: The study may contribute to knowledge on the measures that TVET institutions and other tertiary institutions can develop and implement academic writing practices and BL practices to aid the success of EFAL first-year students. The study was an attempt to provide feedback to academia on the current perspectives and experiences of first-year ESL students at TVET colleges to distinguish areas of limitations with reference to valuable teaching and learning, academic writing and BL practices that compromise quality.

Sebastian Pagel, Michael Jirasek, Leroy Cronin

Chemputation is the process of programming chemical robots to do experiments using a universal symbolic language, but the literature can be error prone and hard to read due to ambiguities. Large Language Models (LLMs) have demonstrated remarkable capabilities in various domains, including natural language processing, robotic control, and more recently, chemistry. Despite significant advancements in standardizing the reporting and collection of synthetic chemistry data, the automatic reproduction of reported syntheses remains a labour-intensive task. In this work, we introduce an LLM-based chemical research agent workflow designed for the automatic validation of synthetic literature procedures. Our workflow can autonomously extract synthetic procedures and analytical data from extensive documents, translate these procedures into universal XDL code, simulate the execution of the procedure in a hardware-specific setup, and ultimately execute the procedure on an XDL-controlled robotic system for synthetic chemistry. This demonstrates the potential of LLM-based workflows for autonomous chemical synthesis with Chemputers. Due to the abstraction of XDL this approach is safe, secure, and scalable since hallucinations will not be chemputable and the XDL can be both verified and encrypted. Unlike previous efforts, which either addressed only a limited portion of the workflow, relied on inflexible hard-coded rules, or lacked validation in physical systems, our approach provides four realistic examples of syntheses directly executed from synthetic literature. We anticipate that our workflow will significantly enhance automation in robotically driven synthetic chemistry research, streamline data extraction, improve the reproducibility, scalability, and safety of synthetic and experimental chemistry.

Lirika Solaa, Youdinghuan Chen, Samantha K. Murphy et al.

Climate change is becoming a widely recognized risk factor of farmer-herder conflict in Africa. Using an 8 year dataset (Jan 2015 to Sep 2022) of detailed weather and terrain data across four African nations, we apply statistical and machine learning methods to analyze pastoral conflict. We test hypotheses linking these variables with pastoral conflict within each country using geospatial and statistical analysis. Complementing this analysis are risk maps automatically updated for decision-makers. Our models estimate which cells have a high likelihood of experiencing pastoral conflict with high predictive accuracy and study the variation of this accuracy with the granularity of the cells.

Matt Earnshaw, Mario Román

We introduce context-free languages of morphisms in monoidal categories, extending recent work on the categorification of context-free languages, and regular languages of string diagrams. Context-free languages of string diagrams include classical context-free languages of words, trees, and hypergraphs, when instantiated over appropriate monoidal categories. Using a contour-splicing adjunction, we prove a representation theorem for context-free languages of string diagrams: every such language arises as the image under a monoidal functor of a regular language of string diagrams.

Jean-Philippe Bernardy, Patrik Jansson

The tensor notation used in several areas of mathematics is a useful one, but it is not widely available to the functional programming community. In a practical sense, the (embedded) domain-specific languages (DSLs) that are currently in use for tensor algebra are either 1. array-oriented languages that do not enforce or take advantage of tensor properties and algebraic structure or 2. follow the categorical structure of tensors but require the programmer to manipulate tensors in an unwieldy point-free notation. A deeper issue is that for tensor calculus, the dominant pedagogical paradigm assumes an audience which is either comfortable with notational liberties which programmers cannot afford, or focus on the applied mathematics of tensors, largely leaving their linguistic aspects (behaviour of variable binding, syntax and semantics, etc.) for the reader to figure out by themselves. This state of affairs is hardly surprising, because, as we highlight, several properties of standard tensor notation are somewhat exotic from the perspective of lambda calculi. We bridge the gap by defining a DSL, embedded in Haskell, whose syntax closely captures the index notation for tensors in wide use in the literature. The semantics of this EDSL is defined in terms of the algebraic structures which define tensors in their full generality. This way, we believe that our EDSL can be used both as a tool for scientific computing, but also as a vehicle to express and present the theory and applications of tensors.

Tebogo J. Rakgogo, Evangeline B. Zungu

The primary focus of this article is the onomastic discrepancies that are considered a blatant disregard of Section 6 (1) of the Constitution of South Africa (Act No. 108 of 1996). The article employs a qualitative research approach where text analysis is used, focusing on constitutional documentation, legislative frameworks on language-related matters and higher education policy documentations, as well as language policy documentation from the selected 10 South African universities that offered Sepedi as a first language or conversational module. It is found that there is a constitutional disregard by competent organisations such as universities, financial institutions, Google and language authorities (such as Pan South African Language Board, Sesotho sa Leboa National Language Body and Sesotho sa Leboa National Lexicography Unit), since most of the policies are inconsistent with Section 6 (1) of the South African Constitution, 1996. In summary, the article records that the policies supersede the Constitution, since the language name ‘Northern Sotho/Sesotho sa Leboa’ is the most used name to refer to the official standard language rather than its counterpart, Sepedi.

Alexander Johnson, Kevin Everson, Vijay Ravi et al.

In this paper, we explore automatic prediction of dialect density of the African American English (AAE) dialect, where dialect density is defined as the percentage of words in an utterance that contain characteristics of the non-standard dialect. We investigate several acoustic and language modeling features, including the commonly used X-vector representation and ComParE feature set, in addition to information extracted from ASR transcripts of the audio files and prosodic information. To address issues of limited labeled data, we use a weakly supervised model to project prosodic and X-vector features into low-dimensional task-relevant representations. An XGBoost model is then used to predict the speaker's dialect density from these features and show which are most significant during inference. We evaluate the utility of these features both alone and in combination for the given task. This work, which does not rely on hand-labeled transcripts, is performed on audio segments from the CORAAL database. We show a significant correlation between our predicted and ground truth dialect density measures for AAE speech in this database and propose this work as a tool for explaining and mitigating bias in speech technology.

Joan Giner-Miguelez, Abel Gómez, Jordi Cabot

Datasets play a central role in the training and evaluation of machine learning (ML) models. But they are also the root cause of many undesired model behaviors, such as biased predictions. To overcome this situation, the ML community is proposing a data-centric cultural shift where data issues are given the attention they deserve, and more standard practices around the gathering and processing of datasets start to be discussed and established. So far, these proposals are mostly high-level guidelines described in natural language and, as such, they are difficult to formalize and apply to particular datasets. In this sense, and inspired by these proposals, we define a new domain-specific language (DSL) to precisely describe machine learning datasets in terms of their structure, data provenance, and social concerns. We believe this DSL will facilitate any ML initiative to leverage and benefit from this data-centric shift in ML (e.g., selecting the most appropriate dataset for a new project or better replicating other ML results). The DSL is implemented as a Visual Studio Code plugin, and it has been published under an open source license.

Matthew Earnshaw, Paweł Sobociński

We introduce regular languages of morphisms in free monoidal categories, with their associated grammars and automata. These subsume the classical theory of regular languages of words and trees, but also open up a much wider class of languages over string diagrams. We use the algebra of monoidal and cartesian restriction categories to investigate the properties of regular monoidal languages, and provide sufficient conditions for their recognizability by deterministic monoidal automata.

Joshua D. White, Anantha Aiyyer, James O. H. Russell

We examined the sensitivity of African easterly waves (AEWs) to elevated terrain over North Africa using a numerical weather prediction model. We formed five ensembles of simulated AEW activity with orographic features independently reduced in four key regions. The ensemble members consisted of 10 consecutive AEW seasons simulated separately. From the ensembles, the southern AEW stormtrack was most sensitive to the reduction of the Ethiopian highlands. Energy budgets showed that diminished diabatic heating associated with precipitating convection was the likely driver of the weaker AEWs. Baroclinic overturning was the dominant pathway for this response. The northern AEW stormtrack was most sensitive to the reduction of the Hoggar and Tibesti mountains. In this case, a reduction in the vertical shear and diminished baroclinic energy conversions from the background state was associated with weaker AEWs. Through terrain reduction, our results provide a view of thermodynamic and dynamic feedback in AEWs that is complementary to what has been shown in past studies.