RESUMEN
The Ebola virus disease (EVD) is an extremely contagious and fatal illness caused by the Ebola virus. Recently, Uganda witnessed an outbreak of EVD, which generated much attention on various social media platforms. To ensure effective communication and implementation of targeted health interventions, it is crucial for stakeholders to comprehend the sentiments expressed in the posts and discussions on these online platforms. In this study, we used deep learning techniques to analyse the sentiments expressed in Ebola-related tweets during the outbreak. We explored the application of three deep learning techniques to classify the sentiments in 8395 tweets as positive, neutral, or negative. The techniques examined included a 6-layer convolutional neural network (CNN), a 6-layer long short-term memory model (LSTM), and an 8-layer Bidirectional Encoder Representations from Transformers (BERT) model. The study found that the BERT model outperformed both the CNN and LSTM-based models across all the evaluation metrics, achieving a remarkable classification accuracy of 95%. These findings confirm the reported effectiveness of Transformer-based architectures in tasks related to natural language processing, such as sentiment analysis.
RESUMEN
Background: Stigma continues to be a major barrier to tuberculosis (TB) control particularly in urban populations. Stigma can influence health seeking behaviour and affect adherence to TB treatment, yet few studies have examined TB related stigma and associated factors in Uganda. This study was therefore conducted to determine the level of stigma and associated factors among TB patients in an urban setting in Kampala, Uganda. Methods: A cross-sectional study was conducted in Makindye division, Kampala among 204 patients with TB aged 18 years and above. Data were collected on socio-demographic, individual patient and HIV/AIDS related factors using an interviewer administered questionnaire. The outcome variable (stigma) was assessed on a four-point Likert scale from the participants' perspective. Stigma scores ranged from 0 to 36 which were summed up and a median stigma score calculated. Individuals with a stigma score equal or greater than the median were categorized as having high stigma. A multivariable logistic regression analysis was performed to determine factors associated with TB stigma. Results: Over half (52%) of the participants were found to have high TB stigma. Knowing someone who had died of TBAOR = 4.42, 95% CI (1.69 - 11.50) and believing that TB and HIV symptoms were similarAOR = 3.05, 95% CI (1.29 - 7.22) were positively associated with high TB stigma. The odds of having high stigma were 79% lower among individuals who had been previously treated for TBAOR = 0.21, 95% CI (0.09 - 0.52). Conclusions: Stigma towards TB was high in this urban population and mainly associated with knowing a person who had died of TB, perception that symptoms of TB are similar to those of HIV/AIDS, and previous TB treatment. Interventions to mitigate TB stigma are needed in urban populations and should also address HIV/AIDS related stigma.