Redirigiendo al acceso original de articulo en 15 segundos...
Inicio  /  Algorithms  /  Vol: 17 Par: 3 (2024)  /  Artículo
ARTÍCULO
TITULO

PDEC: A Framework for Improving Knowledge Graph Reasoning Performance through Predicate Decomposition

Xin Tian and Yuan Meng    

Resumen

The judicious configuration of predicates is a crucial but often overlooked aspect in the field of knowledge graphs. While previous research has primarily focused on the precision of triples in assessing knowledge graph quality, the rationality of predicates has been largely ignored. This paper introduces an innovative approach aimed at enhancing knowledge graph reasoning by addressing the issue of predicate polysemy. Predicate polysemy refers to instances where a predicate possesses multiple meanings, introducing ambiguity into the knowledge graph. We present an adaptable optimization framework that effectively addresses predicate polysemy, thereby enhancing reasoning capabilities within knowledge graphs. Our approach serves as a versatile and generalized framework applicable to any reasoning model, offering a scalable and flexible solution to enhance performance across various domains and applications. Through rigorous experimental evaluations, we demonstrate the effectiveness and adaptability of our methodology, showing significant improvements in knowledge graph reasoning accuracy. Our findings underscore that discerning predicate polysemy is a crucial step towards achieving a more dependable and efficient knowledge graph reasoning process. Even in the age of large language models, the optimization and induction of predicates remain relevant in ensuring interpretable reasoning.

Palabras claves

 Artículos similares

       
 
Leon Kopitar, Iztok Fister, Jr. and Gregor Stiglic    
Introduction: Type 2 diabetes mellitus is a major global health concern, but interpreting machine learning models for diagnosis remains challenging. This study investigates combining association rule mining with advanced natural language processing to im... ver más
Revista: Information

 
Haidi Badr, Nayer Wanas and Magda Fayek    
Unsupervised domain adaptation (UDA) presents a significant challenge in sentiment analysis, especially when faced with differences between source and target domains. This study introduces Weighted Sequential Unsupervised Domain Adaptation (WS-UDA), a no... ver más
Revista: Applied Sciences

 
Razvan Daniel Zota, Ionu? Alexandru Cîmpeanu, Denis Alexandru Dragomir and Mihai Adrian Lungu    
Sustainable development, smart waste management, and circular economy principles are paramount to the significant worldwide trend of smart city-related research and projects. The basic hypothesis of our research is that artificial intelligence (AI)-based... ver más
Revista: Applied Sciences

 
Fátima Trindade Neves, Manuela Aparicio and Miguel de Castro Neto    
In the rapidly evolving landscape of urban development, where smart cities increasingly rely on artificial intelligence (AI) solutions to address complex challenges, using AI to accurately predict real estate prices becomes a multifaceted and crucial tas... ver más
Revista: Applied Sciences

 
Zhengwei Wang, Haitao Gu, Jichao Lang and Lin Xing    
This study verifies the effects of deployment parameters on the safe separation of Autonomous Underwater Vehicles (AUVs) and mission payloads. The initial separation phase is meticulously modeled based on computational fluid dynamics (CFD) simulations em... ver más