Skip to content

saysayani11/MindMatch

Repository files navigation

MindMatch README

Mindmatch is a simple network science implementation to fetch professional details of users and their connections. This is a case of a social network where people are connected to each other via social ties (eg: from work environments, travel friends, colleagues, or simply next door neighbours).

For more details on what MindMatch is, please go over to https://github.com/saysayani11/MindMatch/blob/main/Mindmatch.pdf

TEST DATASET

The dataset used comprises 10000 users with user details randomly generated using Python. The script is divided into four sections:

Section 1: Building the model network Section 2: Preparing the dataset (Part 1) Section 3: Preparing the dataset (Part 2) Section 4: Functions and Analysis.

Check the script for more information.

FUNCTIONS

To test the script, plug in values for any of the functions from the function manual:

  1. know_all_path : Know who connects two users and how
  2. know_shortest_path : Know the shortest possible connection between two users
  3. know_relation : Know how two people are connected
  4. look_for_people_around_me : Who all are common friends among me and user 'n' ?

INPUT ARGUMENTS

1.know_all_path (G,source, target, cutoff) 2. know_shortest_path (G, source, target) 3. know_relation (source,target) 4. look_for_people_around_me (user_id)

MORE DETAILS

G : An Albert-Barabasi graph object of 10000 nodes (users) and starting degree = 2 source : source node (user), datatype = int target : target node (user), datatype = int cutoff : hops cutoff, datatype = int user_id: source node (user), datatype = int

EXAMPLE RUN

Download and run the python script and run each of these functions:

  1. know_all_path (G, 4, 50, 3)

    This means that in the users dataset G comprising 10000 people, we would like to know the number of people that connects user 4 and user 50. "Connections" can happen by way of other people that "know" or "connect" user 4, leading to a "path" till user 50. The fourth argument is the cut-off on the "path" length, i.e, the path between user 4 and 50 should have no more than 3 people.

  2. know_shortest_path (G, 7, 40)

    This function returns the shortest path between two users. From the users dataset of 10000 people, we would like to know the shortest route from user 7 to user 40. In social terms, we wouldl like to know who all are the closest mutuals between user 7 and user 40.

  3. know_relation (5,80)

    Wondering how any two given users can possibly be "connected"? This function does exactly that. Given two users, it can tell you how they are connected (eg: colleagues, same workplace, went to same university and so on)

  4. look_for_people_around_me (50)

    Given any user, cuious to know how you two can be possibly connected? This function gives you a lsit of all the users that connects you to the other user.

About

A "trust-first" network science implementation

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages