To build up a materialized view that perfectly satisfies the need of the specific enterprise it serves is now the biggest challenge especially when it comes to larger and larger scale enterprises as well as more and more complicated and yet necessary socio-economical information. In this paper, we shall develop an Intelligent Materialized VIews Pre-fetching mechanism, also known as an IMVIP, from the characteristics of affinity grouping so as to enhance the efficiency of summary data warehouse querying. The IMVIP mechanism consists of the following two methods: the Apriori-Model association method and the Linear Structure Relation. The Apriori-Model association method explores and deduces the combination of the relations among individual user session. It is especially suitable for applications where the combinations of the relations are to be explored among multi-objective queries made by more than one decision maker. On the other hand, the Linear Structure Relation Model develops a set of principles as to the explorations into the deduced relation combination above with an aim to constructing a series of causal-effect association rules. Thus, we can not only pre-fetch and materialize views that really satisfy the needs of the decision makers so as to enhance the efficiency of summary data warehouse queries but also build up intelligent query paths according to the cause-and-effect association rules in order to attain the goal of providing helpful suggestions for decision-making.
Lee, Chin-Feng and Tsai, Main-Che, "A Design of Intelligent Pre-fetching Materialized Views Mechanism for Enhancing Summary Queries on Data Warehouses" (2001). ICEB 2001 Proceedings. 117.