OLAP Cube-based Graph Approach for Bibliographic Data

There is a growing number of different research fields that are concerned with bibliographic data analysis. In many cases, data of interest can be described as heterogeneous information networks. To explore knowledge from that networks in a multidimensinal way, OLAP (Online Analytical Processing) analysis helps users to access data from different points of views. The ability of OLAP for analyzing classical data is clear. However it must be adapted to provide networked data by considering both nodes and edges. In order to take into account linked data in OLAP on networks, we propose a conceptual graph model to represent bibliographic networks. Then we propose graphs enriched by cubes. Each node and edge of the considered network are described by a cube. It allows the user to quickly analyze the information summarized into cubes. Our proposal also solves the slowly changing dimension problem in OLAP analysis. To illustrate our approach, we integrate three bibliographic databases and a computation process is defined according to user’s needs. Then our implementation shows results on a real data.