Learn R Programming

R/dynetNLAResistance

R/dynetNLAResistance is an R package of anonymization algorithm to resist neighbor label attack in a dynamic network.

Installation

You can install the stable version of R/dynetNLAResistance from CRAN:

install.packages("dynetNLAResistance")

Example

You can create a dynamic network by function make.virtual.dynamic.network. Then group it by function lw.grouping,and anonymize it by function anonymization.

library(dynetNLAResistance)
dynet <- make.virtual.dynamic.network()
dynet.grouped <- lw.grouping(dynet,l = 2, w = 1)
## candidates number:  21  selected id:  V22987 
## Grouping t1 50%
## candidates number:  20  selected id:  V905 
## Grouping t1 100%
## merge.groups
## candidates number:  25  selected id:  V22987 
## Grouping t2 50%
## candidates number:  25  selected id:  V40271 
## Grouping t2 100%
## merge.groups
## candidates number:  30  selected id:  V64428 
## Grouping t3 50%
## candidates number:  29  selected id:  V40271 
## Grouping t3 100%
## merge.groups
## candidates number:  35  selected id:  V64428 
## Grouping t4 50%
## candidates number:  35  selected id:  V40271 
## Grouping t4 100%
## merge.groups
## candidates number:  39  selected id:  V64428 
## Grouping t5 33.33333%
## candidates number:  38  selected id:  V78667 
## Grouping t5 66.66667%
## candidates number:  38  selected id:  V40271 
## Grouping t5 100%
## merge.groups
## candidates number:  40  selected id:  V26750 
## Grouping t6 20%
## candidates number:  39  selected id:  V61271 
## Grouping t6 40%
## candidates number:  39  selected id:  V64428 
## Grouping t6 60%
## candidates number:  38  selected id:  V78667 
## Grouping t6 80%
## candidates number:  39  selected id:  V40271 
## Grouping t6 100%
## merge.groups
## candidates number:  43  selected id:  V26750 
## Grouping t7 16.66667%
## candidates number:  43  selected id:  V35010 
## Grouping t7 33.33333%
## candidates number:  42  selected id:  V22987 
## Grouping t7 50%
## candidates number:  39  selected id:  V78667 
## Grouping t7 66.66667%
## candidates number:  41  selected id:  V40271 
## Grouping t7 83.33333%
## candidates number:  38  selected id:  V8536 
## Grouping t7 100%
## merge.groups
## candidates number:  46  selected id:  V26750 
## Grouping t8 14.28571%
## candidates number:  46  selected id:  V35010 
## Grouping t8 28.57143%
## candidates number:  44  selected id:  V46066 
## Grouping t8 42.85714%
## candidates number:  45  selected id:  V22987 
## Grouping t8 57.14286%
## candidates number:  43  selected id:  V78667 
## Grouping t8 71.42857%
## candidates number:  45  selected id:  V40271 
## Grouping t8 85.71429%
## candidates number:  43  selected id:  V8536 
## Grouping t8 100%
## merge.groups
## candidates number:  49  selected id:  V26750 
## Grouping t9 12.5%
## candidates number:  49  selected id:  V35010 
## Grouping t9 25%
## candidates number:  48  selected id:  V46066 
## Grouping t9 37.5%
## candidates number:  49  selected id:  V22987 
## Grouping t9 50%
## candidates number:  47  selected id:  V78667 
## Grouping t9 62.5%
## candidates number:  46  selected id:  V905 
## Grouping t9 75%
## candidates number:  48  selected id:  V40271 
## Grouping t9 87.5%
## candidates number:  46  selected id:  V8536 
## Grouping t9 100%
## merge.groups
## candidates number:  53  selected id:  V26750 
## Grouping t10 12.5%
## candidates number:  53  selected id:  V35010 
## Grouping t10 25%
## candidates number:  52  selected id:  V46066 
## Grouping t10 37.5%
## candidates number:  53  selected id:  V22987 
## Grouping t10 50%
## candidates number:  51  selected id:  V78667 
## Grouping t10 62.5%
## candidates number:  51  selected id:  V14096 
## Grouping t10 75%
## candidates number:  51  selected id:  V40271 
## Grouping t10 87.5%
## candidates number:  50  selected id:  V8536 
## Grouping t10 100%
## merge.groups
g5.a <- anonymization(dynet.grouped$t5, alpha = 1, beta = 2, gamma = 3)
## Merged group-set's size:  3 
## Vertex number:  43 
## iterating...
## Anonymized nodes with sensitive label:
##  V97632 
## 
## Merged group-set's size:  2 
## Vertex number:  45 
## iterating...
## Anonymized nodes with sensitive label:
##  V97632 V20247 
## 
## Merged group-set's size:  1 
## Vertex number:  53 
## iterating...
## Anonymized nodes with sensitive label:
##  V97632 V20247 V97788

License

MIT

Copy Link

Version

Install

install.packages('dynetNLAResistance')

Monthly Downloads

136

Version

0.1.0

License

MIT + file LICENSE

Maintainer

Last Published

November 25th, 2016

Functions in dynetNLAResistance (0.1.0)

lw.grouping

Generate a grouped dynamic network by lw-grouping algorithm.
cost

Calculate anonymization cost of two nodes.
draw.graph

Draw a graph contains vertexs with sensitive or unsensitive label
anonymize2node

Anonymize two node.
network

Unirected graph: CA-CondMat
make.virtual.dynamic.network

Make a vertex-increasing virtual dynamic network.
anonymization

Anonymize a snapshot of a dynamic network.