This code is supplied as supplementary material for the TVCG paper submission "Influence Maximization with Visual Analytics". The paper presents VAIM, a visual analytics solution for studying information diffusion processes and to tackle the problem of influence maximization.
Paper is published as gold open access, and is available on IEEE Xplore.
VAIM is a client-server application supported by a graph database (Neo4J). The repository has therefore two folders: one with usage and building details for the server and one for the client.
VAIM uses software that is available on all the three major operating systems, specifically:
- GIT and GIT-LFS for downloading the code and data
- Neo4J graph database
- NPM javascript package manager for the client (on Windows, recommended portable installation or the use of a Node version manager)
- OpenJDK java implemenetation (version 11 is preferable since is the one officially supported by Neo4J) and Maven for building and running the server
- GraphViz graph drawing library
System as been tested on Windows 10 & 11 and Ubuntu 22.04, but should be supported on Mac too.
The demo database dumps, given their size, are stored on git using the GIT-Large File Storage (LFS). To make sure those files are correctly downloaded, please check out the repository using git, and not by directly downloading the zip file. If so, the code would still work, but database dumps will malfunction.
Detailed information about the installation of both server and client are available on the respective folders. However, a script for a complete default installation VAIM_install.sh is available for Linux in the scripts folder of this repository. This installs all the required packages and sets up the database. To run it, navigate to the scripts folder and run it as sh VAIM_install.sh. Sudo access will be requested.
For Windows, dependencies need to be individually installed using msi packages or portable installations, making it very difficult to provide an automated installation script. After dependencies have been installed, please refer to the individual server and client readmes to complete the installation.
Once installation is complete, navigate to the scripts folder of VAIM and run VAIM_start.sh script on Linux or VAIM_start.bat on Windows or run the client and server independently with the instructions in the readme files in their respective folders.
A live demo of the system can be accessed here.