This article reviews datasets for COVID-19 detection using ML.
problem Lack of accessible datasets for COVID-19 detection research.
method Analyzed 96 papers on COVID-19 detection from January 2020 to June 2020.
result Identified and summarized datasets used in COVID-19 detection studies.
New method for credible intervals of Covid19 reproduction number.
problem Lack of credibility intervals in existing estimates.
method Combines Langevin Monte Carlo with Proximal operators.
result Produces credible intervals for reproduction number estimates.
Paper analyzes and predicts Covid19 in Romania using neural networks and regime switching.
problem Inaccurate reported numbers and multiple influencing factors in pandemic prediction.
method Three-stage analysis using SIR model refined with neural networks and regime switching.
result Daily estimation of parameters and identification of regime turning points for predictions.
AI model predicts stock prices using social media data and hybrid neural networks.
problem Predicting stock price movements during the COVID-19 pandemic.
method Integrates social media trends and historical stock data using a hybrid CNN-BLSTM framework.
result The proposed framework outperforms traditional models in predicting stock price movements.
Model predicts US COVID-19 deaths with quantile estimates.
problem Predicting US COVID-19 deaths at county level.
method Hybrid machine learning and epidemiological approach, minimizing pinball loss.
result Quantile estimates accurately forecast deaths for different forecast periods.
Epidemiological model updates infection rates in China, US, Italy.
problem Evaluate policy interventions for mitigating the 2019-nCov pandemic.
method Custom SITR model with variational data assimilation for real-time updates.
result Model robust to initial conditions, infers infection numbers and parameters.
TradeR uses RL to execute trades in real markets, minimizing surprise and catastrophe.
problem Minimizing surprise and catastrophe in high-frequency trading.
method Hierarchical RL with energy-based surprise value function.
result TradeR outperforms in abrupt price changes and maintains profitability.
The paper finds that bear markets cause recessions and bull markets cause expansions, with bull markets having a stronger causal effect.
problem Understanding the asymmetric causal relationships between market conditions and economic cycles.
method Asymmetric causality tests using partial sums of positive and negative market components, with bootstrap simulations and leverage adjustments.
result Bear markets cause recessions and bull markets cause expansions, with bull markets having a stronger causal effect.
Young investors, especially students, dominate Indonesian stock exchanges.
problem Investment behavior of young and rookie investors in the stock market.
method Qualitative approach with descriptive analysis and interviews.
result Perception of behavioral control influences investment decisions.
Machine learning model diagnoses COVID-19 from routine blood tests.
problem Difficulty in diagnosing COVID-19 due to inconsistent blood parameter changes.
method Constructed a machine learning model using 5,333 patients with various infections and 160 COVID-19-positive patients.
result Cross-validated AUC of 0.97, sensitivity of 81.9%, specificity of 97.9%.
RKT model improves knowledge tracing by considering exercise relations and student forget behavior.
problem Traditional KT models fail to consider both exercise relations and student forget behavior.
method RKT model uses relation-aware self-attention to incorporate exercise relations and student forget behavior.
result RKT model outperforms state-of-the-art KT methods on real-world datasets.
Study predicts global trade impacts using deep learning during the COVID-19 period.
problem Forecasting global trade impacts during the COVID-19 pandemic.
method Developed a sustainable prediction process using Long-Short Term Memory (LSTM) deep learning model.
result Accurately predicted daily imports and exports for the next 180 days during the pandemic.