Robust query processing for linked data fragments

Linked Data Fragments (LDFs) refer to interfaces that allow for publishing and querying Knowledge Graphs on the Web. These primarily differ in their expressivity exploring different trade-offs when balancing workload between clients servers decentralized SPARQL query processing. To devise efficient plans, typically rely heuristics leverage metadata provided by LDF interface, since obtaining fine-grained statistics from remote sources is a challenging task. However, these are prone potential estimation errors based which can lead inefficient executions with high number of requests, large amounts data transferred, and, consequently, excessive execution times. In this work, we investigate robust processing techniques Fragment address challenges. We first focus plan selection proposing CROP, optimizer explores cost robustness alternative plans. Then, new class adaptive operators: Polymorphic Join Operators. operators adapt join strategy response possible cardinality errors. The results our experimental study show CROP outperforms state-of-the-art plans robustness. second study, how planning approaches benefit polymorphic find they enable more majority cases.