Analyzing Graphs Using Algorithms

You can analyze graphs using basic graph algorithms, graph analysis algorithms, and graph metric algorithms.

The procedure is as follows:

1. Log in to the GES graph editor. For details, see Accessing the GES Graph Editor.
2. In the algorithm library area, you can select an algorithm and set its parameters.

   Algorithm List shows the algorithms supported by GES and Algorithms describes the algorithm details.

   Figure 1 Setting algorithm parameters
   
   

   Algorithms such as PersonalRank, K-hop, and Shortest Path that use the source (vertex ID) and target parameters for query now support querying vertices by property. However, this feature is currently only available for memory edition graphs.

   Figure 2 Querying a vertex
   
3. Run the algorithm by clicking . You can view the query result after the analysis is complete.
   

   Only the results of 500 vertices are displayed due to the size of the result display area. If you want to view the complete query results of global iterative algorithms, such as the PageRank algorithm, you can call the algorithm APIs. For details, see Algorithm APIs.

   Take the sample movie data in the template as an example. The following figure shows the PageRank values.

   Figure 3 Viewing the analysis result
   
4. Adjust the parameters, and run the algorithm again. PageRank value is different this time, but the top ranking does not change.
   Figure 4 Adjusting parameters
   
5. Perform association prediction to obtain the association degree of the two movies. The association degree is 0.029, indicating that only a small group of people have watched both movies.
   Figure 5 Association analysis
   
   Figure 6 Association analysis result