Development of generic CNN deep learning method using feature graph

we propose a method by applying Convolutional Neural Networks (CNN) to non-structured data. CNN has been successful in many fields such as image processing and speech recognition. On the other hand, it was difficult to adapt CNN to n non-structured data such as a csv file with multiple variables. The sequence of the data of the low dimensional grid structure such as the image has a meaning, and the CNN recognizes the order as the feature of the image and processes it. Due to this constraint, CNN could not perform feature recognition on nonstructured data whose sequence can be reordered while leaving the meaning intact. In this work we developed a method to tackle this issue and make CNN applicable by endowering meaning to the sequence of non-structured data, and demonstrated its effectiveness by adding improvements.