Abstract:A reliable knowledge structure is a prerequisite for building effective adaptive learning systems and intelligent tutoring systems. Pursuing an explainable and trustworthy knowledge structure, we propose a method for constructing causal knowledge networks. This approach leverages Bayesian networks as a foundation and incorporates causal relationship analysis to derive a causal network. Additionally, we introduce a dependable knowledge-learning path recommendation technique built upon this framework, improving teaching and learning quality while maintaining transparency in the decision-making process.

What problem does this paper attempt to address?

The main objective of this paper is to construct a network structure that can reveal the causal relationships between knowledge components, in order to enhance the interpretability and credibility of knowledge transmission paths in educational systems. Specifically, the research team proposed a method based on Bayesian networks to construct a causal knowledge network and developed a reliable knowledge learning path recommendation technique on this basis. The core contributions of the paper can be summarized as follows: 1. **Proposing the concept of a causal knowledge network**: To go beyond the existing research that only focuses on the associations (rather than causality) between knowledge components, the authors proposed a new method to construct a network structure that reflects the causal relationships between knowledge components. 2. **Using Bayesian networks to construct the foundational network**: By using Bayesian networks as the foundational model, combined with causal relationship analysis, a foundational network structure that reflects the causal links between knowledge components was constructed. 3. **Intervention and counterfactual experiments to verify causal relationships**: After constructing the foundational network, intervention experiments on specific knowledge components and counterfactual experiments were conducted to further verify and refine these causal relationships. 4. **Knowledge learning path planning**: Based on the constructed causal knowledge network, a method for recommending learning paths was proposed, aiming to plan the most effective learning path according to the actual mastery of the students. Through the above methods, the research aims to improve teaching quality and efficiency, providing support for personalized education. Additionally, the study discusses the challenges and limitations that the proposed method may face and provides an outlook on future research directions.

Enhancing Explainability of Knowledge Learning Paths: Causal Knowledge Networks

Causal discovery using dynamically requested knowledge

The impact of prior knowledge on causal structure learning

CausalKG: Causal Knowledge Graph Explainability using interventional and counterfactual reasoning

Multi-modal Knowledge-aware Reinforcement Learning Network for Explainable Recommendation

Recommendation with Causality Enhanced Natural Language Explanations

Knowledge Enhanced Graph Neural Networks for Explainable Recommendation

Causal Reinforcement Learning for Knowledge Graph Reasoning

Structural Learning of Causal Networks

Explaining the Behavior of Black-Box Prediction Algorithms with Causal Learning

Learning Post-Hoc Causal Explanations for Recommendation

Explainable Reasoning over Knowledge Graphs for Recommendation

Active Learning of Causal Networks with Intervention Experiments and Optimal Designs

Causal Discovery and Knowledge Injection for Contestable Neural Networks (with Appendices)

Reinforced Explainable Knowledge Concept Recommendation in MOOCs

Explainable Reinforcement Learning Through a Causal Lens

Explainable Reinforcement Learning via a Causal World Model

Design of Explainability Module with Experts in the Loop for Visualization and Dynamic Adjustment of Continual Learning

Attention Is Not the Only Choice: Counterfactual Reasoning for Path-Based Explainable Recommendation

KR-GCN: Knowledge-aware Reasoning with Graph Convolution Network for Explainable Recommendation