Explaining Trained Neural Networks with Semantic Web Technologies: First Steps

TL;DR

This paper proposes a conceptual approach using Semantic Web technologies to explain the input-output behavior of trained neural networks, leveraging publicly available structured web data for interpretability.

Contribution

It introduces a novel method that applies Semantic Web technologies to interpret neural networks, demonstrating its feasibility through an experimental proof of concept.

Findings

Semantic Web technologies can effectively explain neural network behavior

Structured web data enhances interpretability of AI models

Proof of concept validates the approach's potential

Abstract

The ever increasing prevalence of publicly available structured data on the World Wide Web enables new applications in a variety of domains. In this paper, we provide a conceptual approach that leverages such data in order to explain the input-output behavior of trained artificial neural networks. We apply existing Semantic Web technologies in order to provide an experimental proof of concept.