Neural Class Expression Synthesis

TL;DR

This paper introduces neural class expression synthesizers that translate positive and negative node sets into class expressions, enabling faster and more accurate explainable node classification in knowledge graphs.

Contribution

It proposes a novel translation-based approach to class expression learning, replacing traditional search methods with neural models like LSTMs, GRUs, and set transformers.

Findings

Synthesizes high-quality class expressions in about one second.

Achieves better F-measures than state-of-the-art on large datasets.

Effective on four benchmark datasets.

Abstract

Many applications require explainable node classification in knowledge graphs. Towards this end, a popular ``white-box'' approach is class expression learning: Given sets of positive and negative nodes, class expressions in description logics are learned that separate positive from negative nodes. Most existing approaches are search-based approaches generating many candidate class expressions and selecting the best one. However, they often take a long time to find suitable class expressions. In this paper, we cast class expression learning as a translation problem and propose a new family of class expression learning approaches which we dub neural class expression synthesizers. Training examples are ``translated'' into class expressions in a fashion akin to machine translation. Consequently, our synthesizers are not subject to the runtime limitations of search-based approaches. We study three instances of this novel family of approaches based on LSTMs, GRUs, and set transformers, respectively. An evaluation of our approach on four benchmark datasets suggests that it can effectively synthesize high-quality class expressions with respect to the input examples in approximately one second on average. Moreover, a comparison to state-of-the-art approaches suggests that we achieve better F-measures on large datasets. For reproducibility purposes, we provide our implementation as well as pretrained models in our public GitHub repository at https://github.com/dice-group/NeuralClassExpressionSynthesis