search

sfu-bigdata / range-driver-tutorials

0 stars 1 forks source link

Data Visualization/Analysis Tutorial #6

Open jillianderson8 opened 3 years ago

jillianderson8 commented 3 years ago

Write the Data Visualization Tutorial

Go through the built-in reporting & visualization functionality (boundary map, bivariate distribution plots, heatmaps) and briefly mention other tools that could be used for other visualizations (seaborn, etc).

I think this should go through the defaults of the functions, but then also explain slight variations. For example, how do we change which variables are being compared in the bivariate distribution plots?

jdpye commented 3 years ago

A really amazing example shared with me just now when I was showing off some of the plots at our Study Hall - https://cdnsciencepub.com/doi/10.1139/cjfas-2020-0078 (see figure 13)

This would be a great single-plot summation of what a given array's expected detection ranges would be. It might be useful for us to build out as part of the spatial visualizations.

jillianderson8 commented 3 years ago

This is a cool paper! Thanks for sharing @jdpye!

Adding an option to plot something like Figure 13 would be pretty easy from my perspective.

The big question I would have is how do we determine an array's expected range. I know that's the big question the range tests are trying to answer and the whole reason we're investigating the impact of environmental factors on detection range, but I'm not entirely sure how we go from our detection rate to a range.

jdpye commented 3 years ago

I think the Brownscombe 2016 paper might have a method we could implement, one that would benefit from the sourcing of environmental data that's being done in the workflows.

https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.13322

jdpye commented 3 years ago

I've also recently seen some 'stoplight' style plots where you grade areas red (most determined-to-be-dominant factors indicate a poor range at the location), yellow (some factors indicate poor range, some indicate good range), or green (all factors determined important to range are favourable in this region). Lots of interesting things happening this week in conference-land!