We introduce AfriEconQA, a specialized benchmark dataset for African economic analysis grounded in a comprehensive corpus of 236 World Bank reports. The task of AfriEconQA is to answer complex economic queries that require high-precision numerical reasoning and temporal disambiguation from specialized institutional documents. The dataset consists of 8,937 curated QA instances, rigorously filtered from a pool of 10018 synthetic questions to ensure high-quality evidence-answer alignment. Each instance is composed of: (1) a question requiring reasoning over economic indicators, (2) the corresponding evidence retrieved from the corpus, (3) a verified ground-truth answer, and (4) source metadata (e.g., URL and publication date) to ensure temporal provenance. AfriEconQA is the first benchmark focused specifically on African economic analysis, providing a unique challenge for Information Retrieval (IR) systems, as the data is largely absent from the pretraining corpora of current Large Language Models (LLMs). We operationalize this dataset through an 11-experiment matrix, benchmarking a zero-shot baseline (GPT-5 Mini) against RAG configurations using GPT-4o and Qwen 32B across five distinct embedding and ranking strategies. Our results demonstrate a severe parametric knowledge gap, where zero-shot models fail to answer over 90 percent of queries, and even state-of-the-art RAG pipelines struggle to achieve high precision. This confirms AfriEconQA as a robust and challenging benchmark for the next generation of domain-specific IR and RAG systems. The AfriEconQA dataset and code will be made publicly available upon publication.
This Table of Contents reflects the print compilation of peer-reviewed articles published in the journal. Each article listed was originally published online under the journal’s open access model and remains individually accessible and citable. This compilation has been created solely for print distribution, reference, and archival purposes. No new research content is introduced. The publisher affirms that all articles included in this compilation have undergone the journal’s standard editorial and peer-review processes.
This bibliometric study aims to analyze the translated literature in the Arab nations that exhibit the highest levels of dynamism spanning from 1979–2012, using the Index Translationum database. Furthermore, it aims to determine the status of the Arabic language in the translation industry by identifying the most frequently translated Arabic works into other languages and the languages with the highest number of translations into Arabic. The study inclusion criteria consider several factors: the source language (Arabic), the target language (Arabic), the Arab nations included in The Index Translationum, and the foreign languages that are most frequently translated to and from Arabic. It is found that despite the rich history and extensive usage of the Arabic language in the Middle East and North Africa, it is ranked 29th among the top 50 languages for translated literature. However, it is noted that the Index, which makes it possible to observe the mapping of translation in the Arab world in accordance with specific parameters, provides no statistics on certain Arab countries beyond 2012. Therefore, the Index does not provide up-to-date information. Further, the ranking of Arab countries in the Index does not seem to align with the official data of certain countries such as Egypt, Lebanon, Saudi Arabia, and Algeria.
Nishat Raihan, Mohammed Latif Siddiq, Joanna C. S. Santos
et al.
Large language models (LLMs) are becoming increasingly better at a wide range of Natural Language Processing tasks (NLP), such as text generation and understanding. Recently, these models have extended their capabilities to coding tasks, bridging the gap between natural languages (NL) and programming languages (PL). Foundational models such as the Generative Pre-trained Transformer (GPT) and LLaMA series have set strong baseline performances in various NL and PL tasks. Additionally, several models have been fine-tuned specifically for code generation, showing significant improvements in code-related applications. Both foundational and fine-tuned models are increasingly used in education, helping students write, debug, and understand code. We present a comprehensive systematic literature review to examine the impact of LLMs in computer science and computer engineering education. We analyze their effectiveness in enhancing the learning experience, supporting personalized education, and aiding educators in curriculum development. We address five research questions to uncover insights into how LLMs contribute to educational outcomes, identify challenges, and suggest directions for future research.
Is it sensical to ascribe psychological predicates to AI systems like chatbots based on large language models (LLMs)? People have intuitively started ascribing emotions or consciousness to social AI ('affective artificial agents'), with consequences that range from love to suicide. The philosophical question of whether such ascriptions are warranted is thus very relevant. This paper advances the argument that LLMs instantiate language users in Ludwig Wittgenstein's sense but that ascribing psychological predicates to these systems remains a functionalist temptation. Social AIs are not full-blown language users, but rather more like Italo Calvino's literature machines. The ideas of LLMs as Wittgensteinian language users and Calvino's literature-producing writing machine are combined. This sheds light on the misguided functionalist temptation inherent in moving from equating the two to the ascription of psychological predicates to social AI. Finally, the framework of mortal computation is used to show that social AIs lack the basic autopoiesis needed for narrative façons de parler and their role in the sensemaking of human (inter)action. Such psychological predicate ascriptions could make sense: the transition 'from quantity to quality' can take place, but its route lies somewhere between life and death, not between affective artifacts and emotion approximation by literature machines.
Translation currently is one of the biggest currency earners globally whose net worth stands in billions of US Dollars. Perhaps, when compared to teaching, translation becomes the second richest single entity in a pool of Language Service Providers. Kiswahili language, in its quest to be among the top languages globally, has embraced translation and increasingly expanding its horizon. This paper therefore purposed to find out the state and quality of online translation tools like Google Translate as used by Kiswahili clients in translation services. The research methodology used was qualitative. A random sampling technique was used to get sample words, phrases, and sentences. The results show that Google Translate has demonstrated excellent translation results when it comes to individual lexemes as compared to phrases and sentences. In some cases, the phrases, especially when used figuratively, have the potential to mislead and give birth to spurious translation. The results of this research will go a long way in helping improve online translation from and into Kiswahili hence not only improving Kiswahili but also opening potentialities of other African languages.
The Information and Communication sector has undoubtedly played a pivotal role in changing the way people live nowadays. Almost every area of our lives is affected by the presence and the use of the new information and communication technologies. In this regard, many researchers' attention has been attracted by the influence or the significant impact of these technologies on economic growth and development. Although the history of South Africa has had some drawbacks that could constitute a big obstacle to the emergence of a successful economic environment, the actual status of the country regarding its economy and the role that it plays in Africa towards the rest of the African countries is a vital example of an emerging economic force in Africa. This paper examines the crucial role that ICT has played and is still playing in the South African economy growth and more specifically the significance of the economic effects of the software industry. It makes use of the framework used by Heavin et al. (2003) to investigate the Irish software industry in order to analyze the impact of endogenous factors -- national, enterprise and individual -- on the software industry and its implication on the economic growth in South Africa.
Mohannad Barakat, Noha Magdy, Jjuuko George William
et al.
Gliomas, the most prevalent primary brain tumors, require precise segmentation for diagnosis and treatment planning. However, this task poses significant challenges, particularly in the African population, were limited access to high-quality imaging data hampers algorithm performance. In this study, we propose an innovative approach combining the Segment Anything Model (SAM) and a voting network for multi-modal glioma segmentation. By fine-tuning SAM with bounding box-guided prompts (SAMBA), we adapt the model to the complexities of African datasets. Our ensemble strategy, utilizing multiple modalities and views, produces a robust consensus segmentation, addressing intra-tumoral heterogeneity. Although the low quality of scans presents difficulties, our methodology has the potential to profoundly impact clinical practice in resource-limited settings such as Africa, improving treatment decisions and advancing neuro-oncology research. Furthermore, successful application to other brain tumor types and lesions in the future holds promise for a broader transformation in neurological imaging, improving healthcare outcomes across all settings. This study was conducted on the Brain Tumor Segmentation (BraTS) Challenge Africa (BraTS-Africa) dataset, which provides a valuable resource for addressing challenges specific to resource-limited settings, particularly the African population, and facilitating the development of effective and more generalizable segmentation algorithms. To illustrate our approach's potential, our experiments on the BraTS-Africa dataset yielded compelling results, with SAM attaining a Dice coefficient of 86.6 for binary segmentation and 60.4 for multi-class segmentation.
Noemi La Bella, Sara Issaoun, Freek Roelofs
et al.
The Event Horizon Telescope (EHT) has recently published the first images of the supermassive black hole at the center of our Galaxy, Sagittarius A* (Sgr A*). Imaging Sgr A* is plagued by two major challenges: variability on short (approximately minutes) timescales and interstellar scattering along our line of sight. While the scattering is well studied, the source variability continues to push the limits of current imaging algorithms. In particular, movie reconstructions are hindered by the sparse and time-variable coverage of the array. In this paper, we study the impact of the planned Africa Millimetre Telescope (AMT, in Namibia) and Canary Islands telescope (CNI) additions to the time-dependent coverage and imaging fidelity of the EHT array. This African array addition to the EHT further increases the eastwest (u, v) coverage and provides a wider time window to perform high-fidelity movie reconstructions of Sgr A*. We generated synthetic observations of Sgr A*'s accretion flow and used dynamical imaging techniques to create movie reconstructions of the source. To test the fidelity of our results, we used one general-relativistic magneto-hydrodynamic model of the accretion flow and jet to represent the quiescent state and one semi-analytic model of an orbiting hotspot to represent the flaring state. We found that the addition of the AMT alone offers a significant increase in the (u, v) coverage, leading to robust averaged images during the first hours of the observating track. Moreover, we show that the combination of two telescopes on the African continent, in Namibia and in the Canary Islands, produces a very sensitive array to reconstruct the variability of Sgr A* on horizon scales. We conclude that the African expansion to the EHT increases the fidelity of high-resolution movie reconstructions of Sgr A* to study gas dynamics near the event horizon.
Yousuf A. Khan, Clarisse Hokia, Jennifer Xu
et al.
The COVID-19 pandemic led to 1.1 million deaths in the United States, despite the explosion of coronavirus research. These new findings are slow to translate to clinical interventions, leading to poorer patient outcomes and unnecessary deaths. One reason is that clinicians, overwhelmed by patients, struggle to keep pace with the rate of new coronavirus literature. A potential solution is developing a tool for evaluating coronavirus literature using large language models (LLMs) -- neural networks that are deployed for natural language processing. LLMs can be used to summarize and extract user-specified information. The greater availability and advancement of LLMs and pre-processed coronavirus literature databases provide the opportunity to assist clinicians in evaluating coronavirus literature through a coronavirus literature specific LLM (covLLM), a tool that directly takes an inputted research article and a user query to return an answer. Using the COVID-19 Open Research Dataset (CORD-19), we produced two datasets: (1) synCovid, which uses a combination of handwritten prompts and synthetic prompts generated using OpenAI, and (2) real abstracts, which contains abstract and title pairs. covLLM was trained with LLaMA 7B as a baseline model to produce three models trained on (1) the Alpaca and synCovid datasets, (2) the synCovid dataset, and (3) the synCovid and real abstract datasets. These models were evaluated by two human evaluators and ChatGPT. Results demonstrate that training covLLM on the synCovid and abstract pairs datasets performs competitively with ChatGPT and outperforms covLLM trained primarily using the Alpaca dataset.
Andrew Renninger, Valentina Marín Maureira, Carmen Cabrera-Arnau
et al.
Transport systems are vulnerable to disruption. This is particularly true in Africa, where there are large areas with few highways and heightened risk of violence. Here we attempt to estimate the costs of violent events on African transport in order to understand the way that it may be limiting integration between regions. In the absence of detailed data on trade or migration, we quantify the cost of violence by relating observed incidents to imputed spatial interaction between cities. We produce indices representing the expected intensity of violent events $μ$ and the expected strength of interaction $ν$ between cities in the African interurban network. We estimate the intensity of conflict in a city and, considering the network of all highways on the continent, use a gravity model to generate flows between pairs of cities. We systematically compare $μ$ to $ν$ and classify areas according to their combined impact and intensity. Results show that certain cities and roads in the network contain outsize risk to Africa's transportation infrastructure. These cities have a high propensity for subsequent violence against civilians, and given their role in the network, they also substantially affect regional connectivity and thus economic integration. According to our model, removing just ten edges due to conflict would require rerouting 32$\%$ of trips. The top 100 edges where violence is likely to happen account for 17$\%$ of all trips. We find that cities with the highest $μ-ν$ risk are typically small and medium size with large degree, meaning they act as hubs. Vulnerable areas tend to be characterised by the presence of terrorist groups like Boko Haram in Nigeria or Al Shabaab in Somalia.
Tobi Olatunji, Tejumade Afonja, Aditya Yadavalli
et al.
Africa has a very low doctor-to-patient ratio. At very busy clinics, doctors could see 30+ patients per day -- a heavy patient burden compared with developed countries -- but productivity tools such as clinical automatic speech recognition (ASR) are lacking for these overworked clinicians. However, clinical ASR is mature, even ubiquitous, in developed nations, and clinician-reported performance of commercial clinical ASR systems is generally satisfactory. Furthermore, the recent performance of general domain ASR is approaching human accuracy. However, several gaps exist. Several publications have highlighted racial bias with speech-to-text algorithms and performance on minority accents lags significantly. To our knowledge, there is no publicly available research or benchmark on accented African clinical ASR, and speech data is non-existent for the majority of African accents. We release AfriSpeech, 200hrs of Pan-African English speech, 67,577 clips from 2,463 unique speakers across 120 indigenous accents from 13 countries for clinical and general domain ASR, a benchmark test set, with publicly available pre-trained models with SOTA performance on the AfriSpeech benchmark.
Ces dernières années, tous les gouvernements du monde ont pris conscience de
l’apparition et de la propagation rapide du coronavirus. En effet, ce thème est
devenu l’affaire de tous puisqu’il a touché tous les pays, ce qui a incité les instances
politiques à mener de vastes campagnes de sensibilisation qui amènent les citoyens
à se faire protéger contre cette pandémie par le biais de tous les moyens de communication. Notre corpus sera donc constitué à partir d’extraits et de titres d’articles
tirés du journal le Monde traitant le sujet du coronavirus. Dans la présente étude,
nous envisageons de relever les stratégies discursives et argumentatives adoptées
par les journalistes pour sensibiliser contre cette pandémie.
Language and Literature, African languages and literature
Through a systematic search of English language peer-reviewed studies, we assess how health professionals and the public, worldwide, perceive the health implications of climate change. Among health professionals, perception that climate change is harming health appears to be high, although self-assessed knowledge is low, and perceived need to learn more is high. Among the public, few North Americans can list any health impacts of climate change, or who is at risk, but appear to view climate change as harmful to health. Among vulnerable publics in Asia and Africa, awareness of increasing health harms due to specific changing climatic conditions is high. Americans across the political and climate change opinion spectra appear receptive to information about the health aspects of climate change, although findings are mixed. Health professionals feel the need to learn more, and the public appears open to learning more, about the health consequences of climate change.
Despite increased understanding of how viral infection is involved in asthma exacerbations, it is less clear which viruses are involved and to what extent they contribute to asthma exacerbations. Here, we sought to determine the prevalence of different respiratory viruses during asthma exacerbations. Systematic computerized searches of the literature up to June 2017 without language limitation were performed. The primary focus was on the prevalence of respiratory viruses, including AdV (adenovirus), BoV (bocavirus), CoV (coronavirus), CMV (cytomegalovirus), EnV (enterovirus), HSV (herpes simplex virus), IfV (influenza virus), MpV (metapneumovirus), PiV (parainfluenzavirus), RV (rhinovirus) and RSV (respiratory syncytial virus) during asthma exacerbations. We also examined the prevalence of viral infection stratified by age, geographic region, type of respiratory secretion, and detection method. Sixty articles were included in the final analysis. During asthma exacerbations, the mean prevalence of AdV, BoV, CoV, CMV, EnV, HSV, IfV, MpV, PiV, RV and RSV was 3.8%, 6.9%, 8.4%, 7.2%, 10.1%, 12.3%, 10.0%, 5.3%, 5.6%, 42.1% and 13.6%, respectively. EnV, MPV, RV and RSV were more prevalent in children, whereas AdV, BoV, CoV, IfV and PiV were more frequently present in adults. RV was the major virus detected globally, except in Africa. RV could be detected in both the upper and lower airway. Polymerase chain reaction was the most sensitive method for detecting viral infection. Our findings indicate the need to develop prophylactic polyvalent or polyvirus (including RV, EnV, IfV and RSV) vaccines that produce herd immunity and reduce the healthcare burden associated with virus-induced asthma exacerbations.
D.J.M. Steijger, Chandrima Chatterjee, W. Groot
et al.
Background: Cost-effectiveness is a tool to maximize health benefits and to improve efficiency in healthcare. However, efficient outcomes are not always the most equitable ones. Distributional cost-effectiveness analysis (DCEA) offers a framework for incorporating equity concerns into cost-effectiveness analysis. Objective: This systematic review aims to outline the challenges and limitations in applying DCEA in healthcare settings. Methods: We searched Medline, Scopus, BASE, APA Psych, and JSTOR databases. We also included Google Scholar. We searched for English-language peer-reviewed academic publications, while books, editorials and commentary papers were excluded. Titles and abstract screening, full-text screening, reference list reviews, and data extraction were performed by the main researcher. Another researcher checked every paper for eligibility. Details, such as study population, disease area, intervention and comparators, costs and health effects, cost-effectiveness findings, equity analysis and effects, and modelling technique, were extracted. Thematic analysis was applied, focusing on challenges, obstacles, and gaps in DCEA. Results: In total, 615 references were identified, of which 18 studies met the inclusion criteria. Most of these studies were published after 2017. DCEA studies were mainly conducted in Europe and Africa and used quality health-adjusted measurements. In the included studies, absolute inequality indices were used more frequently than relative inequality indices. Every stage of the DCEA presented challenges and/or limitations. Conclusion: This review provides an overview of the literature on the DCEA in healthcare as well as the challenges and limitations related to the different steps needed to conduct the analysis. In particular, we found problems with data availability, the relative unfamiliarity of this analysis among policymakers, and challenges in estimating differences among socioeconomic groups.