Optimizing SPARQL Query Processing On Dynamic and Static Data Based on Query Response Requirements Using Materialization

To integrate various Linked Datasets, the data warehousing and the live query processing approaches provide two extremes for the optimized response time and quality respectively. The first approach provides very fast responses but su↵ers from providing low-quality responses because changes of original data are not immediately reflected on materialized data. The second approach provides accurate responses but it is notorious for long response times. A hybrid SPARQL query processor provides a middle ground between two specified extremes by splitting triple patterns of the SPARQL query between live and local processors based on a predetermined coherence threshold specified by the administrator. However, considering quality requirements while splitting the SPARQL query, enables the processor to eliminate the unnecessary live execution and releases resources for other queries and is the main focus of my work. This requires estimating quality of the response provided with the current materialized data, compare it with user requirements and determine the most selective sub-queries which can boost the response quality up to the specified level with least computational complexity. In this work, we discuss the preliminary result for estimating the freshness of materialized data, as one dimension of the quality, by extending cardinality estimation techniques and explain the future plan.