Towards Semantic Assessment of Summarizability in Self-service Business Intelligence

Traditional Business Intelligence solutions allow decision makers to query multidimensional data cubes by using OLAP tools, thus ensuring summarizability, which refers to the possibility of accurately computing aggregation of measures along dimensions. With the advent of the Web of Open Data, new external sources have been used in Self-service Business Intelligence for acquiring more insights through ad-hoc multidimensional open data cubes. However, as these data cubes rely upon unknown external data, decision makers are likely to make meaningless queries that lead to summarizability problems. To overcome this problem, in this paper, we propose a framework that automatically extracts multidimensional elements from SPARQL query logs and creates a knowledge base to detect semantic correctness of summarizability.