NLP-Based Subject with Emotions Joint Analytics for Epidemic Articles

Abstract

For the last couple years, governments and health authorities worldwide have been focused on addressing the Covid-19 pandemic;for example, governments have implemented countermeasures, such as quarantining, pushing vaccine shots to minimize local spread, investigating and analyzing the virus??? characteristics, and conducting epidemiological investigations through patient management and tracers. Therefore, researchers worldwide require funding to achieve these goals. Furthermore, there is a need for documentation to investigate and trace disease characteristics. However, it is time consuming and resource intensive to work with documents comprising many types of unstructured data. Therefore, in this study, natural language processing technology is used to automatically classify these documents. Currently used statistical methods include data cleansing, query modification, sentiment analysis, and clustering. However, owing to limitations with respect to the data, it is necessary to understand how to perform data analysis suitable for medical documents. To solve this problem, this study proposes a robust in-depth mixed with subject and emotion model comprising three modules. The first is a subject and non-linear emotional module, which extracts topics from the data and supplements them with emotional figures. The second is a subject with singular value decomposition in the emotion model, which is a dimensional decomposition module that uses subject analysis and an emotion model. The third involves embedding with singular value decomposition using an emotion module, which is a dimensional decomposition method that uses emotion learning. The accuracy and other model measurements, such as the F1, area under the curve, and recall are evaluated based on an article on Middle East respiratory syndrome. A high F1 score of approximately 91% is achieved. The proposed joint analysis method is expected to provide a better synergistic effect in the dataset.