Hashtag Co-occurrence Network

Question

I would attach the annotations.csv file but the file is too large, instructions are in bold and needs to be in Python programming below:

from google.colab import drive

drive.mount('/content/gdrive')

# Give filepath here to the variable filepath

# filepath = '/content/gdrive/MyDrive/ColabNotebooks/annotations.csv'

filepath = '/content/gdrive/MyDrive/annotations.csv

# Import "pandas" package here

import pandas as pd

import numpy as np

import matplotlib.pyplot as plt        #visualisation

import seaborn as sns                  #Visualisation

%matplotlib inline

Create hashtag co-occurrence network: Extract hashtags from the data. Each hashtag is a node. We put an edge between two nodes (hashtags) if two hashtags appear in a tweet.

For the created graph, print number of nodes and number of edges.

Answer 1

Import required packages

import pandas as pd import networkx as nx

Mount the drive to access the annotations.csv file

from google.colab import drive drive.mount('/content/gdrive')

Define the file path

filepath = '/content/gdrive/MyDrive/annotations.csv'

Load the csv file into pandas dataframe

df = pd.read_csv(filepath)

Extract hashtags from the tweets and create a list

hashtags = [] for text in df['text']: hashtag_list = [tag.strip("#") for tag in text.split() if tag.startswith("#")] hashtags.extend(hashtag_list)

Create a co-occurrence network of hashtags

graph = nx.Graph() for i in range(len(hashtags)): for j in range(i+1, len(hashtags)): if hashtags[i] != hashtags[j]: if graph.has_edge(hashtags[i],hashtags[j]): graph[hashtags[i]][hashtags[j]]['weight'] += 1 else: graph.add_edge(hashtags[i],hashtags[j], weight=1)

Print number of nodes and edges

print("Number of nodes:", len(graph.nodes)) print("Number of edges:", len(graph.edges))