Table of Contents
Description
The application of data visualization to exploratory analysis has proven its effectiveness in different scientific fields. Unfortunately, each discipline suffers from specific problems, making it difficult to apply general solutions. MeLVin is a graphical meta-framework for our MEta Language for Visualization. It was created to facilitate the design of interactive coordinated views applications for visual exploration of data using different technologies. The platform is conceived as an abstraction layer placed over existing data visualization and processing environments, enabling their integration. The data analysis workflow can be seamlessly defined using data-flow diagrams (DFD). Unlike most data-flow diagram-based data mining applications, MeLVin allows users to include visualizations as part of the analysis process and not just as a tool to display final results.
Melvin was designed to fulfill the following requirements:
- Modularity: DFD-based systems provide unique features to formulate a complex problem as a workflow of simple tasks.
- Extensibility, interoperability, reusability: Our framework allows the incorporation of new visualization techniques, data processes, and data types as DFD elements. Expert users can implement DFD nodes using the most suitable technology to solve a particular problem. MeLVin will provide interoperability mechanisms to integrate the node with the rest of the DFD elements. New DFD elements and layouts (applications) can be stored, shared, and reused.
- Easiness of use: Non-expert users can create the DFD by dragging, dropping, and connecting elements into the workspace.
- Expressiveness: MeLVin’s data flow model was designed to not constraint the definition of any possible interaction between data views and processing nodes.
Non-functional requirement:
- Integration effort: The development time must be spent designing data views and not incorporating them into the platform. Our system requires a low number of lines of code to integrate visual embeddings.
- Computational overhead: Adding software layers increases memory usage and computation time. The use of these resources is kept to a minimum.
Screenshots
Videos
Demo and Sources
You can try MeLVin* at: https://gmrv.es/gmrvvis/melvin/app/
GitHub repository: https://github.com/gmrvvis/MeLVin
* You must request your demo account at: cosmin.toader |at| urjc.es
Development Team
Nicusor Cosmin-Toader: cosmin.toader |at| urjc.es
Marcos Garcia-Lorenzo: marcos.garcia |at| urjc.es
Acknowledgments
The research leading to these results has received funding from the following entities: The European Union’s Horizon 2020 Research and Innovation Programme under grant agreement no. HBP604102 (785907 - SGA2), the Spanish Government under grants C080020-09 (Cajal Blue Brain Project, Spanish partner of the Blue Brain Project initiative from EPFL) and TIN2017-83132-C2-1-R (VIANA).