RELISON: A framework for REcommending LInks in SOcial Networks

RELISON is a comprehensive framework for social network analysis and recommendation in Java. This framework allows the analysis of multiple structural properties of a social network graph (from simple ones like the nodes degree or the distance between nodes to more complex ones like betweenness or random walk probabilities), detect the communities in a social network, predict and recommend which links are more likely to appear in the network in the future, and analyze the flow of information that travels through the network.

RELISON has an special focus on the link recommendation (also known as contact recommendation) problem in social networks, i.e. recommending those people in a social network with whom a given user might be interested to connect. Among its main goals, RELISON aims to provide efficient tools for executing and evaluating such approaches, considering not only their accuracy, but also other aspects like their novelty, diversity, and the effects such recommendations have on global properties of the networks (as changes in the structural properties or in the speed or diversity of the information diffusion).

Get started

• Introduction
• Install RELISON
• License
• Contact

Network basics

• Networks
• Properties of a graph
• Creation of empty graphs
• Reading / writing graphs
• Graph manipulation
• Accessing the properties of a network

Social network analysis

• Social network analysis
• Experimental configuration
• Integrate structural metrics in a Java project
• Structural metrics summary

Community detection

• Community detection
• Experimental configuration
• Integrate communities in a Java project
• Community detection algorithms summary

Link recommendation

• Link prediction and recommendation
• Data preparation
• Experimental configuration
• Integrate link recommendation / prediction functionalities in a Java project
• Link prediction/recommendation algorithms summary
• Link prediction/recommendation metrics summary

Information diffusion

• Information diffusion
• Simulation description
• Experimental configuration
• Evaluation of the information diffusion process
• Diffusion protocol summary
• Information diffusion metrics summary

Graph generation

• Graph generation
• Graph generation algorithms