SNoRe: Scalable Unsupervised Learning of Symbolic Node Representations

TL;DR

SNoRe introduces a scalable, interpretable method for learning symbolic node representations in networks, enabling bias detection and explanation of predictions, and demonstrating competitive performance on real-world datasets.

Contribution

The paper presents SNoRe, a novel symbolic node embedding method that is interpretable, scalable, and suitable for sensitive applications like biomedical and user profiling tasks.

Findings

SNoRe achieves competitive results compared to variational autoencoders, node2vec, and LINE.

SNoRe's features enable direct explanation of predictions using SHAP.

The method scales efficiently to large networks.

Abstract

Learning from complex real-life networks is a lively research area, with recent advances in learning information-rich, low-dimensional network node representations. However, state-of-the-art methods are not necessarily interpretable and are therefore not fully applicable to sensitive settings in biomedical or user profiling tasks, where explicit bias detection is highly relevant. The proposed SNoRe (Symbolic Node Representations) algorithm is capable of learning symbolic, human-understandable representations of individual network nodes, based on the similarity of neighborhood hashes which serve as features. SNoRe's interpretable features are suitable for direct explanation of individual predictions, which we demonstrate by coupling it with the widely used instance explanation tool SHAP to obtain nomograms representing the relevance of individual features for a given classification. To our knowledge, this is one of the first such attempts in a structural node embedding setting. In the experimental evaluation on eleven real-life datasets, SNoRe proved to be competitive to strong baselines, such as variational graph autoencoders, node2vec and LINE. The vectorized implementation of SNoRe scales to large networks, making it suitable for contemporary network learning and analysis tasks.