Skip to content

mathematical modeling of affiliation recomendation problem in a typical social network

Notifications You must be signed in to change notification settings

slovnicki/AffiliationRecommendation

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

28 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Affiliation Recomendation

Note that current state of README.md file is relatively poor. Soon it will not be.

Social Network

We built our own social network due to restrictions on existing ones. Everything related to the social netowork is in social-network/. config.h contains basic configuration of network, such as NUMBER_OF_USERS and NUMBER_OF_GROUPS. Change this as you please while testing.

Build

g++ -std=c++11 main.cpp Person.cpp Group.cpp -o m

Run

./m

Octave

Everything related to the work in Octave is in octave/. Run main while in Octave command window to see user by user graph depiction. User by user matrix is saved as S, group by user matrix is saved as A and its transpose is saved as A1.

Graph proximity model

To get the Katz measure of co-occurence matrix AA' call function Katz(A,A1,c), where c is the number of desired group recommendations. To get the Katz measure of combined adjacency matrix C call function KatzC(A,A1,S,lambda,c), where c is the number of desired group recommendations, and lambda is a parametar that controls ratio of the weight of friendship to the weight of group membership (usualy used is lambda=0.2).

Tex

Everything related to the work in tex is in tex/.

About

mathematical modeling of affiliation recomendation problem in a typical social network

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • MATLAB 53.2%
  • C++ 45.9%
  • C 0.9%