An Introduction to the Enterprise-Participant (EP) Data Model
Kevin Houzhi Xu, Yi Sun, Bharat Bhargava · INTERNATIONAL JOURNAL OF NEXT-GENERATION COMPUTING · 2014
The relationships among partial computable functions, e.g., the query If applying a function x to a function y yields a function z ?, are undecidable in general, and therefore we do not have corresponding built-in operators in programming languages or database management systems. Limited to a nite set of nite functions, i.e., nitely many functions with nitely many ordered argument-value pairs, however, the relationships are decidable. The Enterprise-Participant (EP) data model is one that manages data as nite sets of nite functions and that has a set of built-in operators to express the relationships. In this article, we introduce the EP data model. Considering a data model as a recursive language, i.e., its executions always terminate, the EP data model is the most expressive among all contemporary data models including the relational data model and hierarchical data models (e.g., XML), and among all contemporary structured data representations including semi-structured data and graph-based data. Further, it mathematically reaches the limit in database management, i.e., we can use the EP data model to manage arbitrarily computable information. The aim is to consolidate heterogeneous data into a general data structure and to o er more powerful built-in operators for data analysis in cloud computing and big data initiatives.