The scale of decentralization envisioned for the presently centralized web requires querying approaches that can query numerous small data sources instead of a few large ones. Link Traversal-based Query Processing (LTQP) is a promising candidate for querying highly decentralized environments that executes queries with zero knowledge of the queried data and discovers data sources on the fly. However, as the engine does not know in advance what data will be queried, creating an optimized query plan before executing the query is challenging. Presently, LTQP is employed for client-side querying, where one engine instance services a single client. Despite this, current engines do not utilize client-specific engine query usage patterns to implement personalized query optimization algorithms. This paper will describe the proposed research approach for implementing personalized query optimization techniques, such as caching or learned query optimizers, for LTQP. The objective is to improve query optimization algorithms through the analysis of historical query engine usage, instead of depending on additional prior information. Personalized optimization will be based on existing work in SPARQL optimization literature and fundamental database theory, adapted to LTQP, and aimed at repeating their success in reducing query execution time. As a result, query engines will gain the capability to query large decentralized environments, enabling applications to function within this emerging decentralized web landscape.