Collective Classification

Code license

This code is released under BSD v2

Package requirements

See the requirements.txt

Methods implemented

Collective Inference

• Iterative Classification
• Relaxation Labeling
• Gibbs Sampling

Relational Classifiers

• weighted-voting (wvRN)
• naive Bayes (nBC)
• logistic regression

Example usage

See the following gist

Citation

Giselle Zeno and Jennifer Neville. 2016. Investigating the Impact of Graph Structure and Attribute Correlation on Collective Classification Performance. In the 12th International Workshop on Mining and Learning with Graphs (MLG), August 14, 2016, San Francisco, CA, USA.

BibTex

@inproceedings{Zeno2016,
  title = {Investigating the Impact of Graph Structure and Attribute Correlation on Collective Classification Performance},
  booktitle = {the 12th {{International Workshop}} on {{Mining}} and {{Learning}} with {{Graphs}} ({{MLG}})},
  author = {Zeno, Giselle and Neville, Jennifer},
  year = {2016},
  pages = {8},
  publisher = {{ACM, New York, NY, USA}},
  address = {{San Francisco, CA, USA}},
}