search

business-science / correlationfunnel

Speed Up Exploratory Data Analysis (EDA)
https://business-science.github.io/correlationfunnel/
Other
132 stars 25 forks source link

Feature Request for CorrelationFunnel: automate with new explain() Fx #2

Open sfd99 opened 5 years ago

sfd99 commented 5 years ago

Great and very clear stepXstep package tutorial, Matt!.

A time-saving suggestion (if I may): in Step: "Examining the Results" (after Step 3), where you have:

marketing_campaign_correlated_tbl %>% filter(feature %in% c("DURATION", "POUTCOME", "PDAYS", "PREVIOUS", "CONTACT", "HOUSING")) %>% plot_correlation_funnel(interactive = FALSE, limits = c(-0.4, 0.4))

Why not "automatically" generate the results for these top 6 dependent variables?. Easy and useful shortcut! :-)

The new, suggested Function: explain() where the default is: show only the 6 TOP vars,

The user can specify any other # of top vars to show, ie: explain(3) to show the 3 top vars, or explain(+3) to show the 3 top positively-correlated vars, or explain(-3) to show the 3 top negatively-correlated vars

Love to see this explain() Fx in correlationFunnel! :-) THANKS, MATT! great job!.

Sfd99 San Francisco

mdancho84 commented 5 years ago

Thanks for this. I will look into new functionality to quickly hone in on features within a correlation range.