A Hybrid Matheuristic for the Spread of Influence on Social Networks - Complementary Data

Description

This dataset contains complementary data to the paper "A Hybrid Matheuristic for the Spread of Influence on Social Networks" [1], which proposes a matheuristic for combinatorial optimization problems involving the spread of information in social networks. For the computational experiments discussed in that paper, we provide: - Two sets of instances, originally obtained from [2-6]; - The solutions attained by exact and heuristic methods; - The collected results; - The matheuristic source code; The directories "benchmark_*/instances/" contain files that describe the sets of instances. Each instance is associated with a graph containing <n> vertices and <m> edges. The first <m> lines of each file contain: <u> <v> where <u> and <v> identify a pair of vertices that determines an undirected edge. The next line contains <n> integers corresponding to the costs of the vertices. The last line contains <n> integers corresponding to the thresholds of the vertices. The directories "benchmark_*/solutions_*/" contain files describing feasible solutions for the corresponding sets of instances. The first line of each file contains: <s> where <s> is the number of vertices in the target set. Each of the next <s> lines contains: <v> where <v> identifies a target. The last line contains an integer that represents the target set cost. The directory "hmf_source_code/" contains an implementation of the matheuristic framework proposed in [1], namely, HMF. This work was supported by grants from Santander Bank, the Brazilian National Council for Scientific and Technological Development (CNPq), the São Paulo Research Foundation (FAPESP), the Fund for Support to Teaching, Research and Outreach Activities (FAEPEX), and the Coordination for the Improvement of Higher Education Personnel (CAPES), all in Brazil. Caveat: The opinions, hypotheses and conclusions or recommendations expressed in this material are the sole responsibility of the authors and do not necessarily reflect the views of Santander, CNPq, FAPESP, FAEPEX, or CAPES. References [1] F. C. Pereira, P. J. de Rezende, and T. Yunes. A Hybrid Matheuristic for the Spread of Influence on Social Networks. 2024. Submitted. [2] S. Raghavan and R. Zhang. A branch-and-cut approach for the weighted target set selection problem on social networks. 2024. https://doi.org/10.1287/ijoo.2019.0012 [3] J. Leskovec and A. Krevl. SNAP Datasets: Stanford Large Network Dataset Collection. 2024. https://snap.stanford.edu/data [4] R. A. Rossi and N. K. Ahmed. The Network Data Repository with Interactive Graph Analytics and Visualization. 2022. https://networkrepository.com [5] J. Kunegis. KONECT – The Koblenz Network Collection. 2013. http://dl.acm.org/citation.cfm?id=2488173 [6] O. Lesser, L. Tenenboim-Chekina, L. Rokach, and Y. Elovici. Intruder or Welcome Friend: Inferring Group Membership in Online Social Networks. 2013. https://doi.org/10.1007/978-3-642-37210-0_40

Steps to reproduce

1. Modify the first line of the Makefile to configure the path to the Gurobi installation folder on your machine. 2. Open a terminal in the "source_code" directory and run the make command. This will compile the source code and create an executable file named "hmf". 3. To run the program, execute the following command in your terminal: ./hmf -input_path <path_to_instance_folder>/<instance_name>.txt -model <model> where <path_to_instance_folder> denotes the path to the directory containing the instance file, <instance_name> specifies the name of the instance, and <model> indicates the integer programming model used in the matheuristic (either PRY or RAG). At the end of the execution, the program will output a solution in a file named <instance_name>.sol as well as a spreadsheet containing details about the run. The following arguments are optional: -threads : maximum number of threads for Gurobi execution (default is 1) -time_limit : time limit (in seconds) for the local search step in the fourth stage (default is 10) -output_dir : path to the directory where output files will be written (default is "./")

Institutions

Categories

Funding

Licence