Collective Classification in Network Data
Prithviraj Sen, Galileo Mark S. Namata, Mustafa Bilgic et al. · AI Magazine · 2008
Many real‐world applications produce networked data such as the worldwide web (hypertext documents connected through hyperlinks), social networks (such as people connected by friendship links), communication networks (computers connected through communication links), and biological networks (such as protein interaction networks). A recent focus in machine‐learning research has been to extend traditional machine‐learning classification techniques to classify nodes in such networks. In this article, we provide a brief introduction to this area of research and how it has progressed during the past decade. We introduce four of the most widely used inference algorithms for classifying networked data and empirically compare them on both synthetic and real‐world data.
Cited by 3,237 works. doi:10.1609/aimag.v29i3.2157 Open-access PDF
Authors
References
- Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
- Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
- Statistical mechanics of complex networks
- Markov Chain Monte Carlo in Practice
- Fast approximate energy minimization via graph cuts
- Near Shannon limit error-correcting coding and decoding : Turbo-codes
- On the Statistical Analysis of Dirty Pictures
- Factor graphs and the sum-product algorithm
- An Introduction to Variational Methods for Graphical Models
- Advances in Neural Information Processing Systems 14