| 1. |
- Adelani, David Ifeoluwa, et al.
(författare)
-
MasakhaNER: Named Entity Recognition for African Languages
- 2021
-
Ingår i: Transactions of the Association for Computational Linguistics. - : MIT Press. - 2307-387X. ; 9, s. 1116-1131
-
Tidskriftsartikel (refereegranskat)abstract
- We take a step towards addressing the under-representation of the African continent in NLP research by bringing together different stakeholders to create the first large, publicly available, high-quality dataset for named entity recognition (NER) in ten African languages. We detail the characteristics of these languages to help researchers and practitioners better understand the challenges they pose for NER tasks. We analyze our datasets and conduct an extensive empirical evaluation of state-of-the-art methods across both supervised and transfer learning settings. Finally, we release the data, code, and models to inspire future research on African NLP.
|
|
| 2. |
- Adeyemi, Adewale, et al.
(författare)
-
Particulate Matter (PM2.5) Characterization, Air Quality Level and Origin of Air Masses in an Urban Background in Pretoria
- 2022
-
Ingår i: Archives of Environmental Contamination and Toxicology. - : Springer Science and Business Media LLC. - 0090-4341 .- 1432-0703. ; 83:1, s. 77-94
-
Tidskriftsartikel (refereegranskat)abstract
- Several sources have been identified as contributing to the concentration of ambient fine particulate matter, which has been associated to a variety of health issues. The chemical characteristics and sources of trace elements in PM2.5, as well as the air quality index, were investigated in this study. Twenty four-hour fine aerosol particles were collected in an urban area in Pretoria, South Africa, from April 2017 to April 2018. Eighteen trace elements were determined using an XEPOS 5 energy-dispersive X-ray fluorescence (EDXRF) spectrometer, while black and organic carbon were estimated using an optical transmissometer from the samples collected. The HYPLIT model (version 4.9) was used to estimate air mass trajectories. Health risk was calculated by comparing it to the World Health Organization's air quality index (AQI). The overall mean PM2.5 concentration of the collected sample equals 21µg/m3. Majority of PM2.5 exceedances were reported during mid-autumn and winter seasons, as compared to daily WHO guidelines and South African standards. S had the highest concentrations, greater than 1µg/m3. Ni, Se, Br and Sb showed they were extremely enriched, (EF > 10) and suggestive of anthropogenic or non crustal origin The 24-h PM, soot, BC and OC were significantly different by the geographical origin of air masses (p < 0.05). The AQI showed that 70% of the samples showed levels above the AQI range of good and healthy air. The findings include details on the concentration, composition, and potential sources of fine PM2.5, which is essential for policy formulation and mitigation strategies in South Africa’s fight against air pollution.
|
|
| 3. |
- Adeyemi, Adewale, et al.
(författare)
-
Source apportionment of fine atmospheric particles using positive matrix factorization in Pretoria, South Africa
- 2021
-
Ingår i: Environmental Monitoring & Assessment. - : Springer Science and Business Media LLC. - 0167-6369 .- 1573-2959. ; 193:11
-
Tidskriftsartikel (refereegranskat)abstract
- In Pretoria South Africa, we looked into the origins of fine particulate matter (PM2.5), based on one-year sampling campaign carried out between 18 April 2017 to 17 April 2018. The average PM2.5 concentration was 21.1± 15.0 µg/m3 (range 0.7 - 66.8 µg/m3), with winter being the highest and summer being the lowest. The XEPOS 5 Energy dispersive X-ray fluorescence (EDXRF) spectroscopy was used for elemental analysis, and the US EPA PMF 5.0 program was used for source apportionment. The sources identified includes fossil fuel combustion, soil dust, secondary sulphur, vehicle exhaust, road traffic, base metal/pyrometallurgical, coal burning. Coal burning and secondary sulphur were significantly higher in winter and contributed more than 50% of PM2.5 sources. The HYSPLIT model was used to calculate the air mass trajectories (version 4.9). During the one-year research cycle, five transportation clusters were established. North Limpopo (NLP), Eastern Inland (EI), Short-Indian Ocean (SIO), Long-Indian Ocean (LIO) and South Westerly-Atlantic Ocean (SWA). Local and transboundary origin accounted for 85%, while 15% were long-range transport. Due to various anthropogenic activities such as biomass burning and coal mining, NLP clusters were the key source of emissions adding to the city's PM rate. In Pretoria, the main possible source regions of PM2.5 were discovered to be NLP and EI. Effective control strategies designed at reducing secondary sulphur, coal burning, and fossil fuel combustion emissions at Southern African level and local combustion sources would be an important measure to combat the reduction of ambient PM2.5 pollution in Pretoria.
|
|
