The usefulness and ease of use of dashboards are essential elements in supporting interactive queries in data warehousing systems, as they provide the analysts the view of critical business metrics that reflect the business performance. However, on-demand dashboard development presents a challenging task for linking semantic equivalent data facts, discovering structural dependencies between and within data sources. In this paper, we introduce an improved approach for dashboard development that is supported by by-example framework and semantic-based data linkage management. In this Dashboard-by-example (DBE) framework, hypergraph-based techniques are adopted to acquire knowledge from heterogeneous and disparate data into homogeneous ontological clusters and partitions. Moreover, the use of hypergraph-guided data linkage provides multiple perspectives on the knowledge space, supporting interaction model to explore, contextualize and aggregate the data depending on the application needs. The applicability of our approach is discussed in a case study scenario, highlight the flexibility and efficiency in handling on-demand requirements in data warehousing systems.
展开▼
