A conceptual modeling framework for network analytics

Abstract

In this paper we propose a conceptual modeling framework for network analysis applications. Within this framework, a data model called the Network Analytics ER model (NAER) is developed, which enables us to manage and analyze network data in a unified way. In particular, not only data requirements but also query requirements can be captured by the conceptual description of network analysis applications. This unified view provides us a flexible platform to build a number of topology schemas upon the underlying core schema for supporting network analysis queries. We also discuss how the semantics of network analysis queries can be modeled at the conceptual level, and explore three possible application areas of using our analytical framework for network analysis applications: (1) governing semantic integrity, (2) improving analysis efficiency, and (3) supporting network dynamics. We believe that conceptual modeling can play an important role in managing and analyzing network data, and contribute to the development of network analytics.