A general model for automated business diagnosis

Explanation of exceptional values in multi-dimensional business databases

“…We present a method that gives the OLAP analyst explanations for significant decreases or increases in business measures, identified at an aggregated level. Our method for automated diagnosis is based on a generic explanation formalism, as described in Feelders (1993) and Feelders and Daniels (2001). Explanation generation is supported by the two internal structures of the OLAP data cube: the business model and the dimension hierarchies.…”

“…Exception identification is a comparison activity by business analysts, based on the general diagnosis task, as depicted in Figure 1 (Feelders and Daniels 2001). The reference object r represents one element from…”

“…Decision-makers may also apply more abstract normative models in the form of statistical models. In this case the expected behaviour represents the statistically normal case (Feelders and Daniels 2001). We distinguish between two broad classes of statistical models that can be used in an OLAP database:…”

“…Causal influences can appear in two forms: * With the exception of subsection 4.2.4, this section gives an overview of the work presented in (Feelders and Daniels 2001;Heckman 2000;Feelders 1993;Kosy and Wise 1984).…”

“…Overview of theory on explanation diagnosis as developed by Feelders (1993) and Feelders and Daniels (2001). Here explanation is discussed in the context of business model M and based on a generic explanation formalism.…”