Business Context
A social media platform wants to identify tight-knit communities within their user base to target specific group advertisements. Due to the large volume of data, the marketing team needs to find only significant communities (at least 4 members) and wants to stress-test the algorithm's performance on a generated dataset.
About the Set : optNetwork
Network analysis and graph algorithms.
Discover all actions of optNetwork Data Preparation
Generating a larger dataset simulating user connections (circular graph structure to ensure cliques exist) to test filtering capabilities.
Copied!
| 1 | |
| 2 | DATA mycas.social_graph; |
| 3 | DO i=1 to 100; |
| 4 | u1 = 'User'||put(i, z3.); |
| 5 | DO j=1 to 3; |
| 6 | u2 = 'User'||put(mod(i+j, 100)+1, z3.); |
| 7 | OUTPUT; |
| 8 | END; |
| 9 | END; |
| 10 | |
| 11 | RUN; |
| 12 |
Étapes de réalisation
1
Run the clique action filtering for communities with a minimum size of 4 to ignore small, irrelevant groups.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | ACTION optNetwork.clique / links={name='social_graph', vars={from='u1', to='u2'}} minSize=4 maxCliques='ALL' out={name='target_communities', replace=true}; |
| 4 | |
| 5 | RUN; |
| 6 |
2
Check the output table to ensure no clique has fewer than 4 members.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | TABLE.fetch / TABLE='target_communities' to=10; |
| 4 | |
| 5 | RUN; |
| 6 |
Expected Result
The system should filter out all small triangles or pairs and return only cliques with 4 or more nodes. The execution should complete without timeout, demonstrating efficiency on the generated graph.