# We are U19886

## Imports & pre-proceessing

import os

import networkx as nx

import matplotlib.pyplot as plt

from matplotlib import pylab

import numpy as np

import pickle

os.chdir('desktop/ELU 501 data science')

## Loading the graph

G = nx.read_gexf("mediumLinkedin.gexf")

## Loading the data

colleges = {}

locations = {}

employers = {}

with open('mediumCollege.pickle', 'rb') as handle:

colleges = pickle.load(handle)

with open('mediumLocation.pickle', 'rb') as handle:

locations = pickle.load(handle)

with open('mediumEmployer.pickle', 'rb') as handle:

employers = pickle.load(handle)

## Seeking Google employees

google_employees = []

for person in employers:

for employer in employers[person]:

if employer == 'google':

google_employees += [person]

nb_of_google_employees = len(google_employees)

## Registering Google employees informations

register = dict()

for employee in google_employees:

college = None

location = None

if employee in colleges:

college = colleges[employee][0]

if employee in locations:

location = locations[employee][0]

register[employee] = [college, location]

## U19886 information

self = ['NoCollege', locations['U19886'][0]]

## U19886 to Google

# Area

test = False

count = 0

for employee in register:

test = 'rockford illinois area' in register[employee]

if test == True:

count += 1

# Shortest path

shortest_pathIG = {}

for employee in google_employees:

shortest_pathIG[employee] = nx.shortest_path(G, 'U19886', employee)

# who is the nearest google employee ?

min = float('inf')

nearest = ['U19886', min]

for employee in google_employees:

L = len(shortest_pathIG[employee])

if L < min:

min = L

nearest = [employee, L]

## Google to U19886

shortest_pathsGI = {}

for employee in google_employees:

shortest_pathsGI[employee] = list(nx.all_shortest_paths(G, employee, 'U19886'))

nb_paths = {}

for employee in google_employees:

nb_paths[employee] = len(shortest_pathsGI[employee])

##

list_nodes = []

for employee in shortest_pathsGI:

for path in shortest_pathsGI[employee]:

for nodes in path:

if not(nodes in list_nodes):

list_nodes += [nodes]

H = G.subgraph(list_nodes)

##

nodes = list_nodes

k=0

while k < len(nodes):

if nodes[k] == 'U19886':

del nodes[k]

k += 1

##

pos = nx.spring_layout(H, iterations = 1000)

nx.draw_networkx_nodes(H, pos, nodelist = ['U19886'], node_color = 'r', node_size = 20)

nx.draw_networkx_nodes(H, pos, nodelist = nodes, node_color = 'b', node_size = 20)

nx.draw_networkx_nodes(H, pos, nodelist = google_employees, node_color = 'g', node_size = 20)

nx.draw_networkx_edges(H, pos)

plt.axis('off')

plt.show()

# On voit que le no U7091 est un fdp

##

nx.draw_networkx(H, pos, node_size = 20, font_size = 9)

plt.show()

##

nodes1 = list(G.nodes)

k=0

while k < len(nodes1):

if nodes1[k] == 'U7091':

del nodes1[k]

k += 1

P = G.subgraph(nodes1)