Part 4 - Symbolic AI and Knowledge Representation

While deep learning and LLMs dominate modern AI, symbolic AI techniques remain valuable for problems that require explicit reasoning, verifiable logic, and structured knowledge. In this part we explore both classical symbolic AI and practical knowledge representation.

We cover Prolog for logic programming and inference, MiniZinc for constraint satisfaction, the Soar cognitive architecture, and frame-based knowledge representation. We then turn to modern knowledge representation using graph databases, relational databases, and Semantic Web technologies including RDF, SPARQL, and linked data — tools that are widely used in enterprise AI systems today.