Umair Areeba, Masciari Elio
Department of Electrical Engineering and Information Technologies, University of Naples Federico II, Naples 80125, Italy.
Institute for High Performance Computing and Networking (ICAR), National Research Council, Naples, Italy.
Procedia Comput Sci. 2022;203:753-758. doi: 10.1016/j.procs.2022.07.112. Epub 2022 Aug 12.
The whole world is facing health challenges due to wide spread of COVID-19 pandemic. To control the spread of COVID-19, the development of its vaccine is the need of hour. Considering the importance of the vaccines, many industries have put their efforts in vaccine development. The higher immunity against the COVID can be achieved by high intake of the vaccines. Therefore, it is important to analysis the people's behaviour and sentiments towards vaccines. Today is the era of social media, where people mostly share their emotions, experience, or opinions about any trending topic in the form of tweets, comments or posts. In this study, we have used the freely available COVID-19 vaccines dataset and analysed the people reactions on the vaccine campaign using artificial intelligence methods. We used TextBlob() function of python and found out the polarity of the tweets. We applied the BERT model and classify the tweets into negative and positive classes based on their polarity values. The classification results show that BERT has achieved maximum values of precision, recall and F score for both positive and negative sentiment classification.
由于新冠疫情的广泛传播，整个世界都面临着健康挑战。为了控制新冠病毒的传播，研发其疫苗是当务之急。考虑到疫苗的重要性，许多行业都在努力进行疫苗研发。通过大量接种疫苗可以获得更高的新冠免疫力。因此，分析人们对疫苗的行为和态度很重要。如今是社交媒体时代，人们大多以推文、评论或帖子的形式分享他们对任何热门话题的情绪、经历或观点。在本研究中，我们使用了免费可得的新冠疫苗数据集，并使用人工智能方法分析了人们对疫苗接种活动的反应。我们使用了Python的TextBlob()函数，找出了推文的极性。我们应用了BERT模型，并根据推文的极性值将其分为负面和正面类别。分类结果表明，BERT在正面和负面情绪分类方面都取得了最高的精确率、召回率和F分数值。
