Mark Abraham Magumba

Mark Abraham Magumba

Auto-generated profile, based on publicly available information

Mark Abraham Magumba

Technological University Dublin

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 September 2018 in Journal of Big Data
Verified
Journal Article
Open access
PDF file
MMMark Abraham Magumba
Mark Abraham Magumba
PNPeter Nabende
Peter Nabende
EMErnest Mwebaze
Ernest Mwebaze
This paper presents an ontology based deep learning approach for extracting disease names from Twitter messages. The approach relies on simple features obtained via conceptual representations of messages to obtain results that out-perform those from word level models. The significance of this development is that it can potentially reduce the cost of generating named entity recognition models by reducing the cost of annotating training data since ontology creation is a one-time cost as the conceptual level the ontology is meant to be fairly static and reusable. This is of great importance when it comes to social media text like Twitter messages where you have a large, unbounded lexicon with spatial and temporal variations and other inherent biases that make it logistically untenable to annotate a representative amount of text for general purpose models for live applications.


26 October 2021 in Journal of Big Data
Verified
Journal Article
Open access
PDF file
MMMark Abraham Magumba
Mark Abraham Magumba
PNPeter Nabende
Peter Nabende
Twitter and social media as a whole have great potential as a source of disease surveillance data however the general messiness of tweets presents several challenges for standard information extraction methods. Most deployed systems employ approaches that rely on simple keyword matching and do not distinguish between relevant and irrelevant keyword mentions making them susceptible to false positives as a result of the fact that keyword volume can be influenced by several social phenomena that may be unrelated to disease occurrence. Furthermore, most solutions are intended for a single language and those meant for multilingual scenarios do not incorporate semantic context. In this paper we experimentally examine different approaches for classifying text for epidemiological surveillance on the social web in addition we offer a systematic comparison of the impact of different input representations on performance. Specifically we compare continuous representations against one-hot encoding for word-based, class-based (ontology-based) and subword units in the form of byte pair encodings. We also go on to establish the desirable performance characteristics for multi-lingual semantic filtering approaches and offer an in-depth discussion of the implications for end-to-end surveillance.

22 September 2023 in Data in Brief
Verified
Journal Article
Open access
Hasifah NamatovuHasifah Namatovu
Hasifah Namatovu
MMMark Abraham Magumba
Mark Abraham Magumba
TOTonny Justus Oyana
Tonny Justus Oyana
This dataset highlights some of the water quality issues in Uganda. The rationale for collecting the water samples was to test and ascertain the level and source of contamination. A total of one hundred and eighty five samples were collected from sixteen districts. At each water point, a sample was collected using a sterile plastic container, which was pre-rinsed with the water to be sampled. Water samples were drawn from protected and unprotected springs, shallow wells, taps, rain tanks, water reservoirs, open and hand dug wells and boreholes and immediately transported on ice to the National Water Quality Reference Laboratory for analysis. At the laboratory, a BWB flame photometer, Ethylenediamine tetraacetic acid (EDTA)1 titration and gallery plus-thermos fisher discreet analyzer were used to analyze metal, nutrient and anion elements. On-site testing of dissolved oxygen, pH, electrical conductivity and turbidity was done using a water data sonde. This data can be used to draw comparative analyses of water quality issues in rural and urban districts and help in identifying the factors that influence water quality variations. The data can further be used for trend analysis and identifying long-term patterns whilst providing insights into pollution sources and the impact of environmental and climate change. Consequently, mathematical and machine learning models can use this data together with other parameters to predict the changes in water quality which information is essential for policy and decisions making. This data can be used by environmental scientists to draw insights into the health of the aquatic biodiversity; geospatial analysts to ascertain proximal water contaminants; public health specialists to analyze pathogens leading to water-borne diseases; water chemists to study the source and cause of water pollution; data scientists to perform predictive and descriptive analyses; and policy makers to formulate laws and regulations.


21 September 2018 in Online Journal of Public Health Informatics
Verified
Journal Article
Open access
PDF file
MMMark Abraham Magumba
Mark Abraham Magumba
PNPeter Nabende
Peter Nabende
EMErnest Mwebaze
Ernest Mwebaze
The social web has emerged as a dominant information architecture accelerating technology innovation on an unprecedented scale. The utility of these developments to public health use cases like disease surveillance, information dissemination, outbreak prediction and so forth has been widely investigated and variously demonstrated in work spanning several published experimental studies and deployed systems. In this paper we provide an overview of automated disease surveillance efforts based on the social web characterized by their different high level design choices regarding functional aspects like user participation and language parsing approaches. We briefly discuss the technical rationale and practical implications of these different choices in addition to the key limitations associated with these systems within the context of operable disease surveillance. We hope this can offer some technical guidance to multi-disciplinary teams on how best to implement, interpret and evaluate disease surveillance programs based on the social web.