Selected article for: "approach follow and deep learning"

Author: Narinder Singh Punn; Sanjay Kumar Sonbhadra; Sonali Agarwal
Title: COVID-19 Epidemic Analysis using Machine Learning and Deep Learning Algorithms
  • Document date: 2020_4_11
  • ID: 0him5hd2_28
    Snippet: The discussed machine learning and deep learning approaches output the possible number of cases for the next 10 days across the world. Figure 6 illustrates the predicted trend of the COVID-19 using SVR, PR, DNN, and LSTM with worldwide data. Table 1 indicates the RMSE score of the approaches computed against the available number of COVID-19 cases. It is also observed that training of LSTM model is heavily dependent on the deviation in the values,.....
    Document: The discussed machine learning and deep learning approaches output the possible number of cases for the next 10 days across the world. Figure 6 illustrates the predicted trend of the COVID-19 using SVR, PR, DNN, and LSTM with worldwide data. Table 1 indicates the RMSE score of the approaches computed against the available number of COVID-19 cases. It is also observed that training of LSTM model is heavily dependent on the deviation in the values, with the fact that larger the deviation more the time it takes to train. Hence, the number of cases were scaled using minmax scaler to fit the LSTM model and later the predicted cases were rescaled to the original range using invert minmax transform from "sklearn" python library. Among these approaches, the visual representation of the prediction from figure 6 and RMSE score as highlighted in table 1 confirms the PR approach as the best fit to follow the growing trend.

    Search related documents:
    Co phrase search for related documents