Interactive_Algorithm_on_Rule-Based_AI_Production_Systems

Interactive Algorithm on Rule-Based AI Systems: Forward & Backward Chaining with Conflict Resolution AI Production Systems Lab (Python)

Lab Objectives

By the end of this class lab, students will be able to:

Understand the components of a production system (knowledge base, working memory, inference engine). Grounding the theoretical concept for Session 5.

1. Apply pattern matching and conflict resolution.
2. Run forward chaining (data-driven inference)
3. Run backward chaining (goal-driven inference).
4. Experiment with different facts and rules interactively.
5. Visualisation of rule firing order.

This (Lab) system supports Symbolic AI (rule-based reasoning), not Machine Learning. However, using ML, we can simulate the same domain as our rule-based lab, but let the system learn rules automatically using Decision Trees (Supervised Learning), Rule Induction Algorithms and Association Rule Mining (Unsupervised). Python: from sklearn.tree import DecisionTreeClassifier, export_text