Explaining Trained Neural Networks with Semantic Web Technologies: First Steps
By Md Kamruzzaman Sarker, Ning Xie, Derek Doran, Michael L. Raymer, Pascal Hitzler. 2017
69 citations recorded by the source corpus.
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.
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Ugur Kursuncu
- Manas Gaur
- Amit P. Sheth
- Roberto Confalonieri
- Nadia Burkart
- Ludovik Çoba
- Marco F. Huber
- Benedikt Wagner
Citations in the corpus
- Knowledge Infused Learning (K-IL): Towards Deep Incorporation of Knowledge in Deep Learning (2019, background): “(Sarker et al. 2017) presents a conceptual framework for explaining artificial neural networks’ classification behavior using background knowledge on the semantic web.”
- Explainable AI (XAI): Core Ideas, Techniques, and Solutions (2022, methodology): “The symbolic element of NS systems commonly consists of Knowledge Representation and Reasoning techniques, such as ontologies and knowledge graphs [28, 47].”
- A historical perspective of explainable Artificial Intelligence (2020, background): “Sarker, Xie, Doran, Raymer, and Hitzler (2017) propose that methods for explanations should be seen as interactive systems.”
- Deep learning for noise-tolerant RDFS reasoning (2019, background): “[24] builds on top of these approaches by describing an early proof of concept that uses Suggested Upper Merged Ontology (SUMO) [25] as background knowledge and description logic formalism to provide human readable explanations of neural network decisions.”
- A Survey on the Explainability of Supervised Machine Learning (2020, methodology): “ns which follow a structure similar to the human understanding of data are better understandable. Explanations for neural networks try to make the process from the input to the output understandable. Sarker et al. (2017) analyze this relation with the help of ontologies. Their proof of concept descr”