Data driven case base construction for prediction of success of marine operations

Abstract

It is a common situation to have lots of recorded data that you want to use for improving a process in your organization or make use of this data to provide new services or products. Starting with one primary data set we describe a system that enhances this data set to a level such that it can be used by a deep learning system. This deep learning system then creates a model based on this data set, trying to predict operational windows for marine operations. Using this model the system extracts cases for use in a CBR-system aimed at providing operational support. This paper describes the partial implementation and results of this system.