Extending the data warehouse for service provisioning data

Abstract

The last few years, there has been an extensive body of literature in data warehousing applications that primarily focuses on basket-type (transactional) data, common in retail industries. In this paper we focus on service provisioning data, that is data that is recorded internally in an organization for provisioning certain business related tasks. Coupling the recorded data with the underlying process and business-practice(s) that generate them is crucial for end-to-end analysis. Our framework is based on a graph description of the process (called a sketch) that is generating this data. Using this sketch, we formalize a new class of aggregate queries that consolidate data from a part of the process, based on a user defined path expression. We then show how to build a compact, non-redundant collection of summary (aggregate) tables and indices for this new type of queries. We first explore how to select a minimum set of views to answer queries with path-expressions over the given sketch. For queries that also include aggregation, we define two partial orders among the views. The first is used to pick the minimum set of aggregate views to answer any query with no false dismissals, while the second describes an augmented set that allows fewer false positives. Computing a non-materialized aggregate is done through appropriate rewriting of the user query. We describe two indexing schemes that use phantom (non-materialized) aggregate values to expedite query processing. Experimental results show these schemes to perform well on synthetic and real datasets.

Ad-hoc OLAP on Trajectory Data

“…In scientific databases, the same seismic DW [15] can be used by both a scientist and a public administration officer in order to explore seismic activity under a commonly accepted definition of earthquake. Even in latest applications of data warehousing techniques on workflow data [11] the data cube is built on top of well defined facts (service provisions).…”

Optimizing Query Shortcuts in RDF Databases

“…Regardless of the specific model used to store these data graphs (RDF/OWL files, relational databases, etc. ), path expressions require substantial processing; for instance, when using relational databases, multiple join expressions are often involved [3,20,22,28].…”

BIIIG: Enabling business intelligence with integrated instance graphs

“…While BIIIG aims at a general and comprehensive end-toend solution for graph-based data integration and business intelligence, previous related studies mostly focussed on specific aspects. Several graph-based frameworks focus on the transformation of source data into problem-specific [14] [7] or more general [15] graph-based data warehouse models but do not aim at preserving all relationships from the source systems that might become of interest in unexpected ways. A notable exception is DB2SNA [23] that extracts social networks from relational databases and analyzes them, albeit it is not concerned with business analytics.…”

“…Second, all relationships are added as edges to the IIG within a loop over all associations (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29). Analogous to nodes, the association-specific mapping µ is used to query relationships, represented as sets of attribute-value pairs (13).…”