EvalNE: A Framework for Evaluating Network Embeddings on Link Prediction

TL;DR

EvalNE is a Python toolkit that streamlines the evaluation of network embedding methods on link prediction tasks, ensuring consistent and reproducible results through automation and comprehensive metrics.

Contribution

It introduces a flexible, easy-to-use framework for evaluating network embeddings on link prediction, supporting various methods and baselines with minimal coding effort.

Findings

Facilitates reproducible evaluation of network embeddings.

Supports diverse link prediction methods and baselines.

Automates hyper-parameter tuning and validation processes.

Abstract

In this paper we present EvalNE, a Python toolbox for evaluating network embedding methods on link prediction tasks. Link prediction is one of the most popular choices for evaluating the quality of network embeddings. However, the complexity of this task requires a carefully designed evaluation pipeline in order to provide consistent, reproducible and comparable results. EvalNE simplifies this process by providing automation and abstraction of tasks such as hyper-parameter tuning and model validation, edge sampling and negative edge sampling, computation of edge embeddings from node embeddings, and evaluation metrics. The toolbox allows for the evaluation of any off-the-shelf embedding method without the need to write extra code. Moreover, it can also be used for evaluating any other link prediction method, and integrates several link prediction heuristics as baselines.