Handling qualitative conditional preference queries in SPARQL: possibilistic logic approach

Abstract

Purpose The purpose of this paper is to address the need for new approaches in locating items that closely match user preference criteria due to the rise in data volume of knowledge bases resulting from Open Data initiatives. Specifically, the paper focuses on evaluating SPARQL qualitative preference queries over user preferences in SPARQL. Design/methodology/approach The paper outlines a novel approach for handling SPARQL preference queries by representing preferences through symbolic weights using the possibilistic logic (PL) framework. This approach allows for the management of symbolic weights without relying on numerical values, using a partial ordering system instead. The paper compares this approach with numerous other approaches, including those based on skylines, fuzzy sets and conditional preference networks. Findings The paper highlights the advantages of the proposed approach, which enables the representation of preference criteria through symbolic weights and qualitative considerations. This approach offers a more intuitive way to convey preferences and manage rankings. Originality/value The paper demonstrates the usefulness and originality of the proposed SPARQL language in the PL framework. The approach extends SPARQL by incorporating symbolic weights and qualitative preferences.