Chapter 6: Link Prediction

linkpred is a python library designed to provide support to unsupervised link prediction analysis.

In this notebook are introduced some of the main features of the library and an overview of its functionalities.

Note:

• this notebook is purposely not 100% comprehensive, it only discusses the basic things you need to get started.
• LinkPred is developed and maintained by Raf Guns (University of Antwerp)

Table of Contents

1. Installing LinkPred
2. Prediction Workflow
   1. Loading the network from file
   2. Computing the desired unsupervied predictors
   3. Evaluation and comparison of different predictors
3. Conclusions

1. Installing LinkPred (to top)

As a first step, we need to make sure that linkpred is installed and working.

The library can be installed using pip: