An interactive educational website on community detection in complex networks, developed as part of the course Networks and Complex Systems by Cristian Candia, Ph.D. and published through CRiSS Lab. The site combines theory, mathematical intuition, and interactive visualizations to explain how communities emerge in graphs and how different algorithms identify them.
This resource introduces the core ideas behind community detection in network science. It covers both objective-function approaches and probabilistic/generative approaches, moving from classical methods based on modularity optimization to more advanced ideas such as spectral methods and stochastic block models.
The site includes:
- A theory section with conceptual explanations, equations, and algorithmic intuition
- An interactive visualization section to compare methods and graph topologies
- A bilingual interface (Spanish / English)
- A responsive layout for desktop and mobile
The website covers the following topics:
- Introduction to the community detection problem
- Modularity and the modularity matrix
- Greedy modularity optimization
- Spectral methods
- Girvan–Newman
- Louvain
- Leiden
- Infomap
- Label Propagation
- Stochastic Block Models (SBM)
- Robustness and comparison of methods
The goal of this website is to provide a rigorous but intuitive introduction to community detection for students and researchers working in:
- Network science
- Complex systems
- Computational social science
- Data science
- Applied graph analysis
It is designed not only to explain how algorithms work, but also when their assumptions differ, what their limitations are, and why different methods may produce different partitions on the same graph.
- Interactive article format
- Mathematical notation rendered with MathJax
- Dynamic network visualizations
- Responsive user interface
- Language switch between Spanish and English
- Structured navigation across theoretical and algorithmic sections
- Metadata and social preview support for web sharing
This site is built as a static web resource using:
- HTML
- CSS
- Vanilla JavaScript
- SVG for explanatory diagrams and network illustrations
- MathJax for equations
- Google Fonts for typography
A typical structure for deployment is:
.
├── index.html
├── index-en.html
├── assets/
├── site.webmanifest
└── README.md
This website was created as an educational resource for teaching community detection in networks, with an emphasis on combining formal rigor, visual intuition, and algorithmic understanding.