EvoLearner: Learning Description Logics with Evolutionary Algorithms

TL;DR

EvoLearner introduces an evolutionary algorithm to efficiently learn description logic concepts from data, significantly outperforming existing methods especially in handling data properties and initial population generation.

Contribution

The paper presents a novel initialization technique and enhanced data property support for evolutionary learning of description logic concepts, improving performance over state-of-the-art approaches.

Findings

Outperforms state-of-the-art on SML-Bench

Supports data properties with maximized information gain

Effective initialization from positive examples via biased random walks

Abstract

Classifying nodes in knowledge graphs is an important task, e.g., for predicting missing types of entities, predicting which molecules cause cancer, or predicting which drugs are promising treatment candidates. While black-box models often achieve high predictive performance, they are only post-hoc and locally explainable and do not allow the learned model to be easily enriched with domain knowledge. Towards this end, learning description logic concepts from positive and negative examples has been proposed. However, learning such concepts often takes a long time and state-of-the-art approaches provide limited support for literal data values, although they are crucial for many applications. In this paper, we propose EvoLearner - an evolutionary approach to learn concepts in ALCQ(D), which is the attributive language with complement (ALC) paired with qualified cardinality restrictions (Q) and data properties (D). We contribute a novel initialization method for the initial population: starting from positive examples, we perform biased random walks and translate them to description logic concepts. Moreover, we improve support for data properties by maximizing information gain when deciding where to split the data. We show that our approach significantly outperforms the state of the art on the benchmarking framework SML-Bench for structured machine learning. Our ablation study confirms that this is due to our novel initialization method and support for data properties.