Development of a big data application architecture for Navy Manpower, Personnel, Training, and Education

Abstract

Navy Manpower, Personnel, Training, and Education (MPTE) decision makers require improved access to the information obtained from the vast amounts of data contained in a number of disparate databases/data stores in order to make informed decisions and understand second- and third-order effects of those decisions. Toward this end, the effort of this research was two-fold. First, it examined and proposed an end-to-end application architecture for performing analytics for MPTE. Second, it developed a decision tree model to predict retention of post-command aviators, using the Cross-Industry Standard Process for Data Mining (CRISP-DM), in support of one Navy MPTE’s main concerns: retention in post-command aviator community. This research concluded that with the exponential collection and growth of diverse data, there is a need for a combination of Big Data and traditional data warehousing architectures to support analytics at MPTE. The data-mining effort developed a preliminary predictive model for post-command aviation retention and concluded that the number of NOBCs, particularly non-aviation NOBCs, was the most important indicator for predicting retention. Additional data sources particularly those that contain Fitness Reports/Evaluations need to be included in order to improve the accuracy of the model.