CLARK: Closed-loop Learning for Adaptive Reasoning over Knowledge Graphs

Hybrid Knowledge Graph and Large Language Model Architectures for Predictive Analytics

This paper reviews hybrid KG–LLM frameworks for predictive analytics, highlighting graph embeddings, Retrieval-Augmented Generation (RAG), transformer-based reasoning, and contextual embedding fusion to improve prediction accuracy, interpretability, and robustness.

AI-Based Knowledge Graphs for Intelligent Decision Support

Experimental results show that AI-driven knowledge graphs significantly enhance decision accuracy, reduce ambiguity, and improve interpretability, achieving up to 85–92% higher decision efficiency compared to traditional methods.

Hybrid Knowledge Graph and Large Language Model Architectures for Predictive Analytics

This paper presents a comprehensive framework integrating graph embeddings, retrieval-augmented generation (RAG), transformer-based reasoning, attention mechanisms, and contextual embedding fusion to improve prediction accuracy, explainability, and robustness.

Integrating and Reasoning with Data-Induced Information: Knowledge Bases of Axioms and Learned Models

Knowledge-Based Machine Learning Models for Decision Support Systems

Decision Support Systems (DSS) are widely used in healthcare, finance, manufacturing, education, transportation, and public administration to support data-driven decision-making. Traditional DSS based on rule-based expert systems and statistical models often struggle to adapt to dynamic and complex environments. To address these limitations, Knowledge-Based Machine Learning (KBML) integrates machine learning with symbolic knowledge representation techniques such as ontologies, semantic networks, expert rules, and domain constraints. By incorporating prior knowledge into the learning process, KBML enhances reasoning, interpretability, transparency, and predictive performance while reducing training requirements. This paper reviews knowledge-based machine learning approaches for intelligent DSS and examines the integration of knowledge engineering principles with supervised, unsupervised, reinforcement, and ensemble learning methods. The roles of ontologies, rule-based inference, semantic reasoning, and knowledge graphs in improving learning effectiveness are also discussed. A comprehensive DSS framework is proposed, consisting of knowledge acquisition, data preprocessing, feature engineering, knowledge representation, model training, inference generation, and decision recommendation modules. Experimental results demonstrate that knowledge-enhanced models achieve higher accuracy, improved decision consistency, reduced uncertainty, and greater interpretability than conventional machine learning approaches. The study also highlights challenges related to knowledge acquisition, scalability, ontology maintenance, and system integration. Future research directions include explainable AI, deep knowledge graphs, federated learning, cognitive computing, and autonomous reasoning systems.

Neuro-symbolic learning over OWL 2 DL via consequence-based compilation to differentiable circuits

OWL 2 DL ontologies, grounded in the description logic $\mathcal{SROIQ}$, express large knowledge bases in biomedicine and the Semantic Web. Neuro-symbolic (NeSy) learners over description logics either embed the ontology in a continuous space, abandoning classical entailment, or restrict to the Horn fragment $\mathcal{EL}^{++}$, which has a single canonical model. We present Baobab, which compiles a $\mathcal{SROIQ}$ ontology with a finite ABox into a Sentential Decision Diagram (SDD): it saturates a propositional core under a consequence-based calculus and instantiates the remaining $\mathcal{SROIQ}$ features (nominals, number restrictions, and the role axioms) over the active domain. The SDD's evidence-conditioned weighted model count then trains a perception network to recognize real images under partial ABox supervision: on an ontology that exercises every distinctive $\mathcal{SROIQ}$ feature, a CNN learns to read MNIST digits coupled by a successor relation and recovers latent ontology concepts that an independent perception leaves at chance. When the supervision admits several ontology-consistent completions, an independent perception collapses onto one, a reasoning shortcut: we show that a mixture indexed by the query's justifications can represent the calibrated posterior no independent perception can, and that seeding it from the circuit's enumerated completions attains the Bayes-optimal posterior on a real-image MNIST task where single-WMC and learned mixtures (the BEARS-ensemble hypothesis class) do not: to our knowledge the first to characterize and mitigate reasoning shortcuts in a non-Horn description logic. Soundness of the compiler and the representation result are machine-checked in Lean 4. Code is available at https://github.com/bio-ontology-research-group/baobab.