Personalized recommendation of SOLAP queries

Abstract

Spatial data warehouses store enormous amount of complex data; These data are historised and aggregated according to several levels of granularity. In addition, spatial data warehouses store both thematic and spatial data that have specific characteristics such as topology and direction. As matter of fact, extracting interesting information by exploiting spatial datawarehouses could be complex and difficult. Users might ignore what part of the warehouse contains the relevant information and what the next query should be. On the other hand, recommendation is a process that proposes personalized queries according to the user's needs. Developing a recommendation system would facilitate information retrieval in spatial data warehouses. This paper proposes an approach to recommend spatial personalized MDX (Multidimensional Expressions) queries. The approach helps users in the process of exploiting spatial data warehouses and retrieving relevant information by recommending personalized MDX queries. The approach detects implicitly the preferences and needs of SOLAP (Spatial OLAP) users using a spatiosemantic similarity measure. The proposal is described theoretically and validated by experiments.