Understanding the Dynamics of the Stack Overflow Community through Social Network Analysis and Graph Algorithms

TL;DR

This paper analyzes Stack Overflow's community dynamics using social network analysis and graph algorithms, highlighting its role in knowledge sharing and proposing advanced models for traffic forecasting, influence maximization, and link prediction.

Contribution

It introduces novel graph-based and deep learning approaches to understand and enhance online community interactions and information flow.

Findings

Improved traffic flow forecasting with ensemble deep learning.

Efficient multi-objective optimization for influence maximization.

Effective graph convolutional neural network for link prediction.

Abstract

This thesis conducts a focused literature review on online communities, centering on Stack Overflow, employing social network analysis and graph algorithms. It examines the evolving landscape of health information quality within the digital ecosystem, emphasizing the challenges posed and the multifaceted nature of quality. The significance of online communities, notably Stack Overflow, as hubs for social interaction and knowledge sharing is underscored. Proposing advanced approaches, the thesis introduces an ensemble deep learning model for traffic flow forecasting, an efficient multi-objective optimization method for influence maximization, and a graph convolutional neural network-based approach for link prediction.