Most Traveled To and From Stations with Weights
So far, we've only looked at our network with unweighted edges. But our edge weights are actually the number of trips, so it seems logical that we would want to extend our analysis of degrees by adding a weighted degree distribution. This is important because while a balanced degree ratio is important, the item that would need to be rebalanced is bikes. If the weights are the same across all stations, then an unweighted degree ratio would work. But if we want to know how many bikes are actually flowing, we need to consider weights.
The weighted analog to degree distribution is strength. We can calculate this with the strength()
function, which presents a weighted degree distribution based on the weight
attribute of a graph's edges.
This exercise is part of the course
Case Studies: Network Analysis in R
Exercise instructions
- Create a data frame containing the following columns.
trip_out
should contain the"out"
weighted degree (strength) distribution oftrip_g_simp
.trip_in
should contain the"in"
weighted degree distribution.ratio
should contain the ratio of "out" degrees divided by "in" degrees.
- Filter
trip_strng
for rows where bothtrip_out
andtrip_in
are greater than10
. - Plot a histogram of the filtered ratios.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
trip_strng <- data_frame(
# Find the "out" strength distribution
trip_out = strength(___, mode = "___"),
# ... and the "in" strength distribution
trip_in = strength(___, mode = "in"),
# Calculate the ratio of out / in
ratio = ___ / trip_in
)
trip_strng_filtered <- trip_strng %>%
# Filter for rows where trips in and out are both over 10
filter(___ > 10, ___ > 10)
# Plot histogram of filtered ratios
hist(___$ratio)