MayBMS_Mirror/apps/random_graph_demo
Oliver Kennedy 7b2599df3b Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
..
datasets Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
img Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
INSTALL Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
README Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
graph.txt Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
import_data.sh Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
index.php Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
layout.css Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
social.php Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00
style2.css Import from http://maybms.cvs.sourceforge.net/ 2019-01-19 18:15:21 -05:00

README

DESCRIPTION

This application demonstrates the use of probabilistic databases for modelling
random graphs and social networks. We consider graphs in which each edge is
assigned a probability for its existence, and where edges are independent of each other.
The application allows the user to generate a random graph with certain parameters such as
number of nodes and edge probability, or load a social network with probability-weighted edges.

To model such a network as a probabilistic database we store the edge relation as an
uncertain table, where each edge is associated with a boolean random variable. After loading
or generating a network, the user can compute the probability of different patterns
in the network. Some of the currently supported queries are:
- probability for the existence of a triangle (four-clique)
- pairs of nodes within four degrees of separation
- nodes which have at least three neighbors with probability higher than 80%
- nodes which are not directly connected but share at least two neighbors

USE

For installation instructions, see the INSTALL file in the current directory.

To use the application:
-- Generate a random graph by opening http://<localhost>:<port>/random_graph_demo/index.php and filling in the graph parameters.
-- You can upload social network data from a tab-separated file where each line is of the form "from <tab> to <tab> probability". To do this open http://<localhost>:<port>/random_graph_demo/social.php and specify a file to upload.
-- Choose a query to run on the random graph from the ones listed in index.php and social.php.