The image would visualize the selected routes (in our test dataset) on a map of DC. It will highlight the different semantic zone types (in some way). This image would be complemented by a table showing the exact number of routes per zone type, and associated #neighborhoods in each zone type (maybe).
We currently have a table with the two types of grouping (semantic and density based), segregated into 5 and 2 zone type groups respectively. It shows the number of routes and number of label clusters per zone type (Table 1) in the paper. This table can incorporate the number of neighborhoods per zone type (if we decide to add them).
This is based on an earlier team discussion:
The image would visualize the selected routes (in our test dataset) on a map of DC. It will highlight the different semantic zone types (in some way). This image would be complemented by a table showing the exact number of routes per zone type, and associated #neighborhoods in each zone type (maybe).
We currently have a table with the two types of grouping (semantic and density based), segregated into 5 and 2 zone type groups respectively. It shows the number of routes and number of label clusters per zone type (Table 1) in the paper. This table can incorporate the number of neighborhoods per zone type (if we decide to add them).