Sampling complex networks: a critical analysis

• Grant number: 21/05018-1
• Support type: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): August 01, 2021
• Effective date (End): July 31, 2022
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal researcher: Francisco Aparecido Rodrigues
• Grantee: Matheus Victal Cerqueira
• Home Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

Even though complex networks and dynamical systems are receiving growing attention from the scientific community, there are still few well understood descriptive methods for the realization of graph sampling procedures. The network sampling problem is essential for the study of real world systems, given that, more often than not, the real network of interest is not completely known or the system in question is so voluminous that it is computationally intractable. Some studies have approached the topic investigating different sampling techniques for networks and applying analytical and empirical methods, although, this field of study still lacks of systematic methodologies that could enable a more procedural analysis of connected data sampling plans, besides the fact that it is a field with numerous open questions because of the inerent complexity of the problem of preserving a networks fundamental characteristics from de obtention of a subgraph. This project has exploratory and investigative nature and has the objective to analyze the use of multivariate statistical analysis and machine learning with the purpose of investigating, in a critical manner, network sampling methods properties with a structured and methodological proposal, looking to obtain different approaches for the study of questions like the comparison of sampling techniques, metrics of interest estimation and inference in complex networks topology. By virtue of the broad project proposal, it is expected to obtain observations that will be useful in the future for the investigation of related topics. (AU)