Abstract
In this paper we introduce a framework for probabilistic reasoning on knowledge graphs. The framework leverages the notion of probabilistic knowledge graphs (PKGs), a dedicated probabilistic graphical model, as well as Soft Vadalog, a specific language for knowledge representation and reasoning on such model. We illustrate PKGs, the language and the general problem of probabilistic reasoning, providing approximate algorithmic tools to make it feasible and efficient. This work—a short version of our recent contribution to the International Joint Conference on Rules and Reasoning 2020—aims at making our results available to the broader statistical community.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
