Visualize the MOOC learner features with MOOC-Learner-Visualized (MLV)
Entry point is
autorun.py. Configuration is done with
config/*yml, see e.g.
As a backend, run MOOC-Learner-Curated only once, run MOOC-Learner-Quantified only when necessary by command line then
run MOOC-Learner-Visualized (the
MOOC-Learner-Visualized serves as an interface to visualize features populated in the
moocdb. It contains three parts:
- Fetching: which fetches feature columns from feature tables
- Processing: provides several processing functions on data columns, such as filtering, mapping, and statistics calculation. Finally, it performs an inner join on relevant feature rows to form a data frame for rendering
- Rendering: provides an interface to configure and render interactive and static plots. The template engine can
as adding a new dictionary describing the configuration form and write up a new drawing function, which takes a
pandasdataframe and the all form input as parameters and return a
bokehfigure instance. Currently we have scatter plot, histogram as templates for interactive plots and scatter matrix as the template for static plots. For interactive plots, the user can select or filter a specific feature column with a slider.