Dynamic graph in a symbolic data framework: An account of the causal relation using COVID-19 reports and some reflections on the financial world

Dual attention-based sequential auto-encoder for Covid-19 outbreak forecasting: A case study in Vietnam

“…Similar to that, recently Chatterjee, A. et al ( Chatterjee, Gerdes, & Martinez, 2020 ) studied on using multiple LSTM-based architectures to efficiently preserve the dynamic temporal information from reported Covid-19 spreading data to conduct accurate predictions. Also related to Covid-19 time series based data evaluation and learning, Nascimento et al ( Nascimento et al, 2021 ) recently propped a novel dynamic graph-based analysis technique with multi-regression dynamic model (MDM) approach. It supports to find relationships between time series routine Covid-19 reported data and financial market trends.…”

“…It is the results of fast person-to-person transmission ( Fidan and Yuksel, 2021 , Pitchaimani and Devi, 2021 ) of the Covid-19 delta variant as well as shortage of suitable strategies for preventing the spreads of virus among groups of confirmed Covid-19 cases to other healthy ones. In addition, the shortage of proper data modelling and short/long-term Covid-19 outbreak forecasting solutions also led to challenges for the governments to effectively manage as well as plan for social resource optimization ( Nascimento et al, 2021 ).The accurate pandemic forecasting mechanism also supports for the governments to properly impose suitable policies to simultaneously deal with the expansion of Covid-19 as well as ensure the social/economic stability ( Miao, Last, & Litvak, 2022 ). Due to the severe influences of this pandemic in multiple social aspects, it is considered as necessary for building data analysis systems which support to capture the spreading temporal patterns of this pandemic.…”