| Title: | Plot Halfcircle Diagram |
|---|---|
| Description: | There are growing concerns on flow data in diverse fields including trade, migration, knowledge diffusion, disease spread, and transportation. The package is an effective visual support to learn the pattern of flow which is called halfcircle diagram. The flow between two nodes placed on the center line of a circle is represented using a half circle drawn from the origin to the destination in a clockwise direction. Through changing the order of nodes, the halfcircle diagram enables users to examine the complex relationship between bidirectional flow and each potential determinants. Furthermore, the halfmeancenter function, which calculates (un) weighted mean center of half circles, makes the comparison easier. |
| Authors: | Sohyun Park, Ningchuan Xiao |
| Maintainer: | Sohyun Park <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.1.0 |
| Built: | 2025-01-19 03:44:48 UTC |
| Source: | https://github.com/cran/halfcircle |
A dataset containing trade data between countries and traded volumes for each 4 category.
data(ex_flow)data(ex_flow)
A data frame with 10866 rows and 6 variables:
name of exporting country
name of importing country
volume of land associated with trading vegetables, in ha
volume of land associated with trading fruits, in ha
volume of land associated with trading wheats, in ha
volume of land associated with trading soybeans, in ha
A dataset containing 154 countries who participate in the trade and related attributes.
data(ex_node)data(ex_node)
A data frame with 154 rows and 8 variables:
name of exporting country
longitude of the center of a country
latitude of the center of a country
total number of population
Gross Domestic Product per capita, in dollar
total volume land for cultivation use, in ha
total volume of usable water, in cubic meter
5 levels by income
halfcircle function draws flows between nodes creating halfcircle diagram.
halfcircle(flow, node, dir = "horizontal", circle.col = "lightgray", circle.trans = 0.5, flow.col = "black", flow.trans = 0.5, flow.width = "proportional", node.color = "black", node.size = 0.1, node.pch = 20, node.trans = 0.7, label = node[, c(1)], label.size = 0.5, label.col = "black", label.gap = 0.1)halfcircle(flow, node, dir = "horizontal", circle.col = "lightgray", circle.trans = 0.5, flow.col = "black", flow.trans = 0.5, flow.width = "proportional", node.color = "black", node.size = 0.1, node.pch = 20, node.trans = 0.7, label = node[, c(1)], label.size = 0.5, label.col = "black", label.gap = 0.1)
flow |
a dataframe which is to draw half-circles. The data should be in the form of an edge list containing node of origin, node of destination, and magnitude of the flow on the first three columns. |
node |
a dataframe which contains names of node on the first column. Nodes on the center line of a circle are drawn by the order of the data. Every node presented in flow data must be contained. |
dir |
if 'horizontal' (the default), nodes are drawn along the X-axis. If 'vertical', nodes are drawn along the Y-axis. |
circle.col |
color of background circle |
circle.trans |
transparency of color of background circle |
flow.col |
flow color. flow.col can be a list of color vectors, the vectors are then used per flow. |
flow.trans |
transparency of color of flows |
flow.width |
width of flows. if 'proportional' (the default), each width is calculated to be proportional to the maximum volume of flows. Maximum width is set to be 10. Otherwise, a list of width vectors can be used per flow. |
node.color |
node color. It can be a list of color vectors, and the vectors are then used per node. |
node.size |
node size |
node.pch |
node type. see ?points for more options. |
node.trans |
transparency of color of flows |
label |
the first column of node, names, is represented (the default). a list of vector an be used per node. if NULL, no label is drawn. |
label.size |
label size |
label.col |
label color |
label.gap |
gap between the node and the respective label |
This function is a low-level graphical function, and you will create a halfcircle diagram. To create the diagram, nodes are placed as a set of points on a straight line segment in the center of a circle. The flow between two nodes is represented using a half cicle drawn from the origin to the destination in a clockwise direction. It is virtually drawn on xy-coordinates where both x and y range from -1 to 1. Flows between the same nodes are not drawn.
Sohyun Park <[email protected]>, Ningchuan Xiao
Xiao and Chun (2009) <doi:10.1559/152304009788188763>
# load flow data data(ex_flow) flow <- ex_flow[,c(1,2,3)] # select veget column as volume flow <- subset(flow,flow$vegetable>5000) data(ex_node) # load node data node <- ex_node[c(order(-ex_node$gdpc)),] # sort nodes in descending order of gdpc values halfcircle(flow, node, dir="vertical", circle.col="gray", flow.col="black",label=NULL) # legend max <- max(flow[,c(3)]); median <- median(flow[,c(3)]); min <- min(flow[,c(3)]) max_w <- 10; median_w <- round(10*median/max); min_w <- round(10*min/max) legend(x=-1.2, y=-0.8, legend=c(paste(round(max)), paste(round(median)), paste(round(min))), lty=1, lwd=c(max_w, median_w, min_w), cex=0.7) # customize colors node$color <- c("#22abcb","#4eb6ad","#86c388","#adcd6c","#dad84f")[node$income_level] flow2 <- data.frame(flow, node[match(flow[,"O"], node[,"country"]),]) halfcircle(flow2, node, dir="vertical", flow.col=flow2$color, node.color=node$color, label=NULL) # highlight one node flow3 <- flow flow3$color <- "gray" flow3$color[flow3$O=="China"|flow3$D=="China"] <- "blue" flow3 <- flow3[c(order(flow3$color,decreasing=TRUE)),] node$label <- "" node$label[node$country=="China"] <- "China" halfcircle(flow3, node, dir="vertical", flow.col=flow3$color, label=node$label, label.size=0.7)# load flow data data(ex_flow) flow <- ex_flow[,c(1,2,3)] # select veget column as volume flow <- subset(flow,flow$vegetable>5000) data(ex_node) # load node data node <- ex_node[c(order(-ex_node$gdpc)),] # sort nodes in descending order of gdpc values halfcircle(flow, node, dir="vertical", circle.col="gray", flow.col="black",label=NULL) # legend max <- max(flow[,c(3)]); median <- median(flow[,c(3)]); min <- min(flow[,c(3)]) max_w <- 10; median_w <- round(10*median/max); min_w <- round(10*min/max) legend(x=-1.2, y=-0.8, legend=c(paste(round(max)), paste(round(median)), paste(round(min))), lty=1, lwd=c(max_w, median_w, min_w), cex=0.7) # customize colors node$color <- c("#22abcb","#4eb6ad","#86c388","#adcd6c","#dad84f")[node$income_level] flow2 <- data.frame(flow, node[match(flow[,"O"], node[,"country"]),]) halfcircle(flow2, node, dir="vertical", flow.col=flow2$color, node.color=node$color, label=NULL) # highlight one node flow3 <- flow flow3$color <- "gray" flow3$color[flow3$O=="China"|flow3$D=="China"] <- "blue" flow3 <- flow3[c(order(flow3$color,decreasing=TRUE)),] node$label <- "" node$label[node$country=="China"] <- "China" halfcircle(flow3, node, dir="vertical", flow.col=flow3$color, label=node$label, label.size=0.7)
Calculate average values of flows and plot them
halfmeancenter(flow, node, dir = "horizontal")halfmeancenter(flow, node, dir = "horizontal")
flow |
a dataframe which is to draw half-circles. The data should consist of node of origin, node of destination, and magnitude of the flow on the first three columns. |
node |
a dataframe which contains names of node on the first column. Nodes on the center line of a circle are drawn by the order of the data. |
dir |
if 'horizontal', nodes are drawn along the X-axis. If 'vertical', nodes are drawn along the Y-axis. |
This function is to get values of mean centers and average radius of flows. One of values of mean centers is weighted by the magnitude of flow and the other one is unweighted. If flows are normally distributed or all combinations of flows between nodes are made, the mean center should be located in the center of a circle, that is (0,0) on the xy-coordinates, and average radius should be 0.5. If the mean center fall in a certain quadrant, a user can evaluate the skewedness.
A list containing calculated average values c(x-coordinate of weighted mean center, y-coordinate of weighted mean center, weighted average radius,x-coordinate of unweighted mean center, y-coordinate of unweighted mean center, unweighted average radius)
Sohyun Park <[email protected]>, Ningchuan Xiao
data(ex_flow) flow <- subset(ex_flow, ex_flow$veget>5000) data(ex_node) node <- ex_node[c(order(-ex_node$gdpc)),] halfmeancenter(flow, node, dir="vertical")data(ex_flow) flow <- subset(ex_flow, ex_flow$veget>5000) data(ex_node) node <- ex_node[c(order(-ex_node$gdpc)),] halfmeancenter(flow, node, dir="vertical")