Integration Maximization for the Agent Placement Problem

Paper: Assigning Agents to Increase Network-Based Neighborhood Diversity
Full version (contains complete proofs of theorems): [https://arxiv.org/pdf/2301.02876.pdf]

Directory Overview

1. algorithm: contains all the code
2. network: contains some example networks

To compile the code

Required software: C++11

Steps:

1. cd algorithm/
2. g++ -std=c++11 -O3 heuristic.cpp heuristic.h other_function.cpp other_function.h graph.h IoA.cpp -o IoA

To run the code: ./IoA network_name algorithm_id k

• network_name:
  • gnp_1000_10 : gnp graph, n = 1000, avg_deg = 10
  • pl_1000_10 : power-law graph, n = 1000, avg_deg = 10
  • google_plus_subnetwork: a subgraph of google_plus
• algorithm_id:
  • 1: local improvement
  • 2: greedy
• k: the number of type-1 agents (should not be larger than n/2)

Example 1: run local improvement on google_plus_subnetwork with k = 100

• ./IoA google_plus_subnetwork 1 100

Example 2: run greedy on gnp_1000_10 with k = 100

• ./IoA gnp_1000_10 2 100