Advances in Network Analysis and Modeling

The field of network analysis is rapidly evolving, with a growing focus on integrating machine learning and statistical physics methods to improve community detection and network modeling. Recent research has highlighted the importance of considering local separators and node-pair similarities in community detection, leading to more accurate and efficient algorithms. Additionally, there is a growing interest in developing application-agnostic methods for estimating video quality of experience (QoE) metrics from encrypted traffic, which has significant implications for network operators. The development of new algorithms and methodologies for network analysis, such as clique annealing and multidimensional scaling, is also underway. Noteworthy papers in this area include those that propose innovative approaches to community detection, such as CLANN, which integrates crystallization kinetics into the optimization process, and those that demonstrate the effectiveness of machine learning in estimating QoE metrics from encrypted traffic. Furthermore, research on network sampling and simplification is providing new insights into the structural characteristics of networks and the most effective methods for analyzing them.

Sources

Machine Learning Informed by Micro and Mesoscopic Statistical Physics Methods for Community Detection

When Machine Learning Meets Importance Sampling: A More Efficient Rare Event Estimation Approach

Visualization Tasks for Unlabelled Graphs

How Local Separators Shape Community Structure in Large Networks

Video QoE Metrics from Encrypted Traffic: Application-agnostic Methodology

Are Widely Known Findings Easier to Retract?

New Recipe for Semi-supervised Community Detection: Clique Annealing under Crystallization Kinetics

A Comparative and Measurement-Based Study on Real-Time Network KPI Extraction Methods for 5G and Beyond Applications

Adaptive continuity-preserving simplification of street networks

Learning Isometric Embeddings of Road Networks using Multidimensional Scaling

Network Sampling: An Overview and Comparative Analysis

Built with on top of