Predictive Analytics of Hyper-Connected Collaborative Network

Abstract

The foundation of the infrastructure of a collaborative network for ubiquitous connectivity will employ hyper-connected technologies in smart and sustainable cities. Typically, there are millions of items for processing and analytics on the massive generated data. The predictive analytics are indispensable for such volumes of which there are many drifts in data structures and contents. In order to make better decisions and future planning of ubiquity, a model, and correspondence implementation are designed and developed. It brings decision-making to the expected boundary of collaboration for different performance indexes. The selected method finds cause-and-effect between data to predict the optimum responses to incoming events. The core of approach focuses on Event-Condition-Action rules to build decision trees, which helps further planning. The method can summarize complexity via effective recommended decisions to local experts and analysts.