CharPDM: Characterized Prompt-oriented Document Melioration (for SEO) – “Seize your text!”

Abstract: In this research we developed LLM-based bots to optimize text ranking on search engines using white hat SEO techniques. These bots interact with a black box ranker to adjust content for improved visibility, prioritizing relevance and quality for both the source document and user queries. Through a structured prompt engineering process, we evaluate and fine-tune bots, selecting models that best balance SEO performance with content integrity. The resulting framework enables effective, ethical ranking improvements while preserving high standards of textual relevance.