Skip to content

User Manual

Jump to bottom
itsnottheend-Jason edited this page Dec 10, 2022 · 4 revisions

User Manual
1. Introductions
This is a demo project focusing on Covid19 virus for the program ‘Data Processing and Visualization’ based on dashboard environment.
The purpose of this project is to perform data exploration, filtering, pre-processing, visualization, basic analysis and prediction, to finally answer the question ‘how to fight coronavirus’ on a data base.

2. HMI Layouts
This dashboard project includes 3 pages, which can be selected on the right-top zone
- Page ‘Explore’: to visualize data.
- Page ‘Analyse’: to perform analysis.
- Page ‘Predict’: to forcast future status of the virus.
image

3. Functions
The specific functions provided by each page of dashboard are explained as follows:
1) To visualize and compare spatial-temporal features of covid on Page ‘Explore’
Select countries:
image

Select the features to show:
image

Select the aggregation method:
image

For this dropdown, more choice are available:

- filter by continents
image

- filter by other data groups, such as Income, education level, HDI, etc
image

The data are visualized with three connected graphs: selected countries (spatial), the data-time(temporal) graph, and a data-selection window using to screen the scope of spatial-temporal to be visualized.
image
imageimage

2) To analyze potential correlations between confirmed cases (death cases) and possibleanalytical factors on Page ‘Analyse’
Select the targe attributes to be investigated such as New Cases, and the focused factors that might be related to it, such as: Mask Usage and Median Age
image

The coefficients of relevance between the target attributes and all other factors are calculated with a lasso method.
image

The data is presented with each sample in the following graph:
image

And the feature-to-feather relationship is plotted in the following graph, in which a bubble plot will present the info of the second features that might need to be investigated.
image

Also a button to shift the two graph (samples-plot and relationship-plot), in case a more clear view is expected.
image

3) To predicte for COVID-19 evolution in the next 3 months on Page ‘Predict’
Select the target country, the target feature to be predicted, and the relevant features that might make contribute to it(Which can be deduced from the page ‘Analyse’)
image

More features can be selected to the understanding of the user is available:
image

And by clicking on the button ‘Run Prediction Model’, the result will be shown.
image

4. Appendix
1) Datasource: https://ourworldindata.org/coronavirus
2) Toolchain: Python(3.10.x), Dash
3) Server adress: http://10.24.92.68
4) For more info, please contact: xiangmis@stud.ntnu.no

Clone this wiki locally