Polygons To Points

This tool converts a GeoJSON, TopoJSON or shapefile polygon file to the "Nodes" CSV file with longitude and latitude coordinates of polygons centroids. For a shapefile choose one .zip, or select the .shp, .shx, .dbf and .prj files together. Click "Submit" to download the file to your browser's download folder.

Each polygon is replaced by one point at its centre, and the points are written to a CSV with the polygon attributes plus x and y.

Flow Matrix to Flow List

This tool converts a CSV file containing a flow matrix into a flow list CSV file with columns for origin, destination, and volume. Select the orientation of the matrix and click "Submit" to download the flow CSV file to your browser's download folder.

An origin-by-destination matrix becomes one row per origin and destination pair with its volume; zero cells are left out.



Flow Matrix Orientation




Regionalization alpha

What it does. A map with hundreds of flows is hard to read. This tool groups places into regions by community detection: it finds groups of places that exchange much more flow with each other than expected, by maximizing modularity. Each community becomes one region, and the flows are added up between regions, so the main patterns stand out with far fewer flow lines. Everything runs in your browser.

Methods. ScLeiden (the default) and ScLouvain are the spatially constrained versions of the Leiden and Louvain algorithms (Wang, Wang & Onega 2021): every region is one contiguous area, and regions smaller than a minimum size are merged into a neighbour. Leiden and Louvain use the flows only, so a community can be spread over several separate areas.

1. Load the node and flow files in the Nodes and Flows tabs and choose their fields. Optionally load region polygons (Regions tab) whose IDs match the node IDs: they give exact boundaries and neighbours.

2. Settings.

Method
Resolution higher = more, smaller regions; lower = fewer, larger ones
Runs Random seed
Randomness θ Leiden refinement; 0.01 as in Wang et al. (0 = always the best merge)
Minimum region size measured as
Region boundaries
Region node

Region node: the point that stands for a region. The flow-weighted centre of a spread-out or curved region can fall outside it, even inside another region; the default then uses the member place with the smallest flow-weighted distance to all other members, which is always inside the region.

Flow Coordinates to Nodes

What it does. Some flow files describe each flow by the coordinates of where it starts and where it ends (for example o_x, o_y, d_x, d_y, volume) instead of by place names or IDs. FlowMapper needs two linked files: a Nodes file that lists every location once with an ID, and a Flows file that says which node each flow goes from and to. This tool finds every distinct location in your flow file, gives it an ID, and creates both files for you.

A flow table with start and end coordinates is split into a nodes table with one row per location and a flows table that refers to node IDs.

1. Choose your flow CSV file. Coordinates should be longitude (x) and latitude (y) in decimal degrees.

Normalize Flows

Flow normalization helps remove the effect of total flow or population size by calculating modularity, i.e., the difference between the actual flow and the expected volume of flow for each pair of locations (nodes). We use the adjusted flow volume formula below to calculate the expected number of flows between locations.

For more details in modularity function please refer to Guo (2009).

Observed flow compared with expected flow for two pairs; modularity is observed minus expected, positive when there is more flow than expected.

normalize flows

1. Select your flow file with columns: origin, destination, and volume, then click "Submit".

2. Once processed, your file will be downloaded into your browser's download folder. It will have columns for origin, destination, modularity, rawvolume, and expectation. The modularity column reflects the difference between observed (rawvolume) and expected flow (expectation) using the given formula.

3. Modularity can be negative or positive. Negative values indicate fewer observed flows, while positive values suggest more observed flows than expected between origins and destinations.

Calculate Node Measures

Node measures calculated from flow data facilitate the creation of custom choropleth base map and node symbol map to provide context into the importance of nodes in the spatial network.

These measures currently include gross (total) flow, inflow, outflow, net flow, entropy, Gini coefficients, and degree measures. Koylu & Guo (2013).

For node B with incoming flows 30 and 20 and an outgoing flow of 40: inflow 50, outflow 40, total flow 90, net flow plus 10, net-flow ratio 0.11.

1. Select your flow file with columns: origin, destination, and volume, then click "Submit".

2. Once processed, node measures file will be downloaded into your browser's download folder in a csv format that contains the nodeids that exist in the flow file and the basic node measures listed above.

3. You could join the node measure file to the choropleth or node symbol map to customize the symbology to reflect the selected measure.

Test Server Connection (beta)

This is a testing tool to connect to the backend Java code, which will provide more complicated analysis tools for pre-processing your data before visualization.

Only for testing use! please input your file below:

Testing result:

Welcome
TEST