Combinging Code and Visual Programming for Data Driven Applications

Graphical programming promises to make complex systems more accessible by visual abstraction and rich interaction. Machine Learning systems, particularly those using Deep Neural Networks, are such complex software systems.

Still, graphical programming is only seldomly used and many developers prefer traditional, text-based editors for their flexiblity.

The goal of this thesis to bridge this divide by combinding graphical and textual programming in a hybrid editor for the data driven development domain.

This thesis will include the following tasks:

• Development of a text/graphical hybrid programming environment that allows the creation of data driven software.
• Evaluation of this editor to determine its success and whether it adresses the concern of developeers with graphical programming.

You will summarize the development and evaluation process and its results in a written thesis and present them to an audience.