ABSTRACT
After the emergence of the Covid-19 virus, pharmaceutical companies began making vaccines against this virus. Peoples' reactions towards vaccines varies between acceptance and rejection. Information about these reactions can be found in social media which has become the largest and best source of users' opinions on a specific topic nowadays. One of the most important social media through which this data can be collected is Twitter. It is important to analyze people's opinions about these vaccines to find out the percentage of supporters and opponents of vaccines. Sentiments analysis can be used to analyze people's opinions. In this paper, we proposed a hybrid deep learning model to analyze user sentiment towards the COVID-19 vaccine. The contributions of our work are to adopt an efficient-designed model by combines Convolutional Neural Network (CNN), which has the capability to extract features, and Long Short-Term Memory (LSTM), which can monitor and study long-term dependencies between words. And provide the proposed network topology setting that contributed in producing high performance in sentiment analysis of the COVID-19 vaccine tweets. Extensive experiments have been conducted on a data set of 13,190 tweets. The results proved that the proposed model with the proposed topology setting outperformed the other machine learning models. © 2022 Little Lion Scientific.
ABSTRACT
Aims: The impact of the COVID-19 pandemic spans all aspects of life. This study aimed to investigate the mental health situation of Jordanian university students during the COVID-19 pandemic. Instrument & Methods: This cross-sectional study on 1000 university students from April to May 2020. A web-based survey that investigates students' psychological distress and anxiety was conducted. Google Form was used to create the survey, and it was published using Facebook and WhatsApp applications over university students' groups. SPSS 19 software was used for analysis. Nonparametric tests (Mann-Whitney and Kruskal-Wallis) were used to examine the significant associations between psychological distress and anxiety;an ordinal regression analysis was also performed. Findings: Of the 1000 students who filled the questionnaire, 39.3% were male, and 60.7% were female. The Mean..SD age of the student was 22..3.8 years old. 42.1% suffer from distress, and 72.6% suffer from anxiety. Furthermore, male gender and family income stability were protective factors against psychological distress and anxiety. Regions (Irbid, Balqa, Jerash, Ajloun, Alzarqa, Tafila, Amman, Aqaba, Karak, Maan) were considered as a risk factor.