Academic Research
Our academic research endeavours epitomise the highest standards of scholarly inquiry in Data Science and Artificial Intelligence. Through rigorous methodologies and peer-reviewed publications, we relentlessly push the boundaries of knowledge, uncovering novel insights and driving advancements at the cutting edge of our field.
Some of Our Award-winning Academic Research Papers
Geo-Semantics Analysis Of Environmental Disasters In Nigeria Using National Print Media Data For Disaster Management
This research paper was submitted to the International Conference on Learning Representations – ICLR 2025
Untapping African Music With Ai: Investigating Music Source Separation Models For Low-Resource African Languages
This research paper was submitted to the International Conference on Learning Representations – ICLR 2025
AI In Yorub` A´ Stem Education For Early Childhood Learning: An Evaluation On Translation Quality And Context
Evaluating Speech To Text For Children With Speech Impairment
This research paper was submitted to the International Conference on Learning Representations – ICLR 2025
Bridging The Language Gap: Evaluating Machine Translation For Animal Health In Lowresource Settings
This research paper was submitted to the International Conference on Learning Representations – ICLR 2025
How Effective Are AI Models In Translating English Scientific Texts To Nigerian Pidgin: A Low-Resource Language?
This research paper was submitted to the International Conference on Learning Representations – ICLR 2025
Neural Machine Translators (NMTs) as Efficient Forward and Backward Arabic Transliterators
This research paper was submitted to the 2024 Conference on Neural Information Processing Systems – NeurIPS 2024
The Influence Of Geo-Location Mentions In The Projection Of A Music
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Geo-Semantic Analysis Of Medical Research Trends In Nigeria
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Leveraging Geo-Nlp For Enhanced Antiretroviral Drug Distribution In Nigeria: Insights From Social Media And News Data
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Prediction Of The Ability And Motivation To Adopt Reproductive Health Behavioural Change Using Anonymized Customer Center Audio Data
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Geo-parsing and Geo-Visualization of Road Traffic Crash Incident Locations from Print Media for Emergency Response and Planning
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024 and
Deep Learning Indaba 2024
Tangalenlp: Building Po Tangle To English Parallel Corpora And Machine Translation Of The Tangle (Tangale) Language
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Enhancing Transformer Models For Igbo Language Processing:A Critical Comparative Study
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Afrihg: News Headline Generation For African Languages
This paper introduces AfriHG—a news headline generation dataset created by
combining from XLSum and MasakhaNEWS datasets focusing on 16 languages
widely spoken by Africa. We experimented with two seq2eq models (mT5-
base and AfriTeVa V2), and Aya-101 LLM.
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Contextual Evaluation Of LLM’s Performance On Primary Education Science Learning Contents In The Yoruba Language
This research critically assesses the ability of LLMs, including ChatGPT-3.5, Gemini, and PaLM 2, to comprehend and generate contextually
relevant science education content in Yoruba.
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Comparative Study Of LLMs For Personal Financial Decision In Low Resource Language
This study seeks to investigate the accuracy
of LLM models in responding to basic day-to-day financial questions both in the
English and Yoruba languages. The result shows that ChatGPT4.0 outperformed
ChatGPT3.5 and Bard(LaMDA) in all three phases. T
This research paper was submitted to the International Conference on Learning Representations – ICLR 2024
Using Advanced Analytics Of Social Media Content To Understand Rape In Nigeria
Rape is a worldwide epidemic which stands on the same pedestal with the offence of murder and has been largely characterized by a culture of impunity. The definition of rape is inconsistent between governmental health organizations, law enforcement, health providers, and legal professions. It has varied historically and culturally.
AI Class Monitor
This work explores how Machine Learning-powered cameras in a classroom can help teachers to recognize and assess their pupils’ engagement levels and classroom behavioural patterns with the goal of intervening when students need motivation and assistance.
AFRIGAN: African Fashion Style Generator using Generative Adversarial Networks(GANs)
Afrocentric fashion images suitable for machine learning tasks are not well represented in open datasets.In this work we present AFRIGAN, a generative adversarial model for contemporary african fashion images.
Semantic Enrichment of Nigerian Pidgin English for Contextual Sentiment Classification
Nigerian English adaptation, Pidgin, has evolved over the years through multilanguage code switching, code mixing and linguistic adaptation. While Pidgin preserves many of the words in the normal English language corpus, both in spelling and pronunciation, the fundamental meaning of these words have changed significantly.
A Comparative Analysis of Semi-Supervised & Self-supervised Classification for Labeling Tweets about Police Brutality
Social media has proven to be influential in social justice advocacy. In 2020, thousands of Nigerians and allies used the EndSARS hashtag to protest police brutality in Nigeria.
A Practical Framework for Building & Managing Digital Crowdsourcing Communities
Digital communities for data crowdsourcing are here to stay. This work presents our tested framework for activation, maintenance, and long-term engagement of digital crowdsourcing communities for data collection.
A spaCy model for Medical Named Entity Recognition(NER) in Nigeria Social Media
Africa is known for its rich and diverse culture. Africans are flamboyant and fashion is an expression of our shared heritage as a people.Traditional African fashion is rich, colourful and vibrant,and allows for adaptable customization.
AFRIFASHION1600: A Contemporary African Fashion Dataset for Computer Vision
Fashion is everywhere. It is one of the main ways that humans present themselves to others. It signals what individuals want to communicate about their sexuality, wealth, professionalism, subcultural and political allegiances. African fashion is as diverse and dynamic as the continent and the people who live there.
AFRIFASHION40000 : A GAN generated dataset for African Fashion
We present AFRIFASHION40000, an openly available dataset of African fashion images generated using Generative Adversarial Networks (GANs). This work explores the practical application of Artificial Intelligence to contemporary African art for new designs, image data synthesis, and representation in computer vision tasks.
AFRIRAZER - A Deep learning Model to remove background & Skin from Traditional African Fashion Images
AFRIRAZER is a deep learning model for background and skin removal in African fashion images. While the concept of image segmentation using deep learning for fashion is not new, traditional African images are not well represented in the common fashion training images.
AI-powered understanding of Family Planning Behavioral Change using the Fogg Model
In this work, we share our methodology and findings from applying named entity recognition (NER) using machine learning, to identify behavioural patterns in transcribed family planning client call centre data in Nigeria based on the Fogg’s model.
Artificial Intelligence for Pharmacovigilance in Nigerian Social Media
Under-reporting of suspected adverse drug reactions (ADRs) to public health facilities is one of the challenges facing pharmacovigilance in Nigeria. The advent of social media has opened up a new wave of communication where people can openly discuss a range of topics and they can choose to use pseudonyms.
COVNLP: A Multisource COVID-19 Dataset for Natural Language Processing
We propose COVNLP, a novel dataset for Natural language processing tasks.
The openly available dataset consists of 3,199 de-identified peer-to-peer messages shared across different channels like Whatsapp, SMS and Social media channels from volunteers during the COVID-19 pandemic in Nigeria.
EDUSTT: In-Domain Speech Recognition for Nigerian Accented Educational Content in English
English Automatic Speech Recognition systems are trained on regular speech, therefore they may struggle to perform well on accented and domain-specific speech.
Federated Grassroots Community Model for Catalyzing Artificial Intelligence for Common Good
We are proposing a methodology for developing Artificial Intelligence solutions that adopt a grassroots federated community model involving a rare collaboration between problem solvers, end-users, and other stakeholders in identifying problems and conceptualization and development of Artificial Intelligence solutions.
NaijaNER: Comprehensive Named Entity Recognition for 5 Nigerian Languages
We present our findings on Named Entity Recognition for 5 Nigerian Languages (Nigerian English, Nigerian Pidgin English, Igbo, Yoruba and Hausa). These languages are considered low-resourced, and very little openly available Natural Language Processing work has been done in most of them.
Problems and Solutions Documentation Template for Machine Learning Competitions
Democratizing access to Artificial Intelligence and truly utilizing it for the common good requires multi-stakeholder AI competitions focussed on real and prevalent problems with the potential for large scale impact and that promotes and ensures the explainability, reproducibility, contextualization and incremental enhancements of solutions.
Real-Time Crowdsourcing of Health Data in Low-income Countries
Malaria is one of the leading causes of high morbidity and mortality in Nigeria despite various policy interventions to frontally address the menace. While a national malaria policy agenda exists on the use of Artemisinin-based antimalarial as the first-line drug of choice for the treatment.