DIRE: A Scalable RDFSE Reasoner For Time-Varying RDF Graphs
Name
Bruno Rucy Carneiro Alves de Lima
Abstract
The core tenet of Symbolic Artificial Intelligence is the concept of machine reasoning: the act of deducting knowledge, new or old, from already-existing machine-readable facts. In spite of the ice age that Symbolic AI has been enduring, certain industry fields that heavily depend on knowledge graphs, such as the medical one, not only still make usage of machine reasoning, but do so with ever- increasing amounts of data that change at an even greater rate. New techniques are needed in order to reincarnate machine reasoning into the age of big data. Our contribution lies in the development of a reasoning engine that is built upon a streaming and distributed computational model that has first-class support for calculating fixpoints and handling both the addition and removal of data efficiently, one of the most complicated challenges that permeate the field. The state of the art is significantly outperformed in symmetric performance between additions and deletions, while at the same time being distributed.
Graduation Thesis language
English
Graduation Thesis type
Master - Computer Science
Supervisor(s)
Victor Henrique Cabral Pinheiro
Defence year
2022