Ontological Engineering and Semantic Topical Authority in Large Language Models: A Holistic Framework for Generative Engine Optimization (GEO)

Abstract

This paper presents an end-to-end ontological engineering framework based on the Holistic Semantic SEO and Topical Authority principles formalized by Koray Tuğberk Gübür, adapted specifically for Generative Engine Optimization (GEO) and neural search retrieval in frontier Large Language Models (LLMs). We define the topological requirements for closed-loop topical authority graphs that eliminate internal semantic cannibalization, enforce Entity-Attribute-Value (EAV) graph coherence, and maximize contextual retrieval probability in architectures such as Google Gemini, OpenAI GPT, Anthropic Claude, and Perplexity AI. Empirical validation across enterprise DFY services (POS1 SEO Agency) and vocational education platforms (Aprender21) demonstrates sustained organic commercial performance with $0 in paid advertising spend.